JOURNAL OF CLINICAL MICROBIOLOGY, Sept. 1980, p. 406-408 Vol. 12, No. 3 0095-I I37/80/09-0406/03$02.00/0

Isolation of fetus subsp. jejuni from Migratory Waterfowl NANCY A. W. LUECHTEFELD,'* MARTIN J. BLASER,2t L. BARTH RELLER,2 AND WEN-LAN L. WANG' Microbiology Laboratory, Denver Veterans Administration Medical Center, Denver, Colorado 80220,12 and Department of Pathology' and Division of Infectious Diseases, Department ofMedicine,2 University of Colorado School ofMedicine, Denver, Colorado 80262 Since the sources from which humans acquire Campylobacter enteritis are only partially known, we studied the frequency of carriage of Campylobacter fetus subsp. jejuni in migratory waterfowl. Cecal contents of various species of wild ducks were cultured on selective media that contained to inhibit normal flora. Thirty-five percent of the 445 ducks cultured harbored C. fetus subsp. jejuni. Migratory waterfowl are yet another reservoir for this enteric pathogen and may be of public health importance for humans in the contamina- tion of water or when used as food. Campylobacter fetus subsp. jejuni has been from these two sources. recognized only recently as a human enteric The cecal and cloacal swabs were inoculated on a pathogen. Selective methods of culture have en- selective plate medium that consisted of brucella agar abied isolation of this in 3 to 14% of cases base (BBL Microbiology Systems, Cockeysville, Md.), agent 591 sheep blood, and the following antimicrobials, per of acute diarrheal illness in humans (1, 4, 13, 15). liter: amphotericin B, 2 mg; cephalothin, 15 mg; poly- Despite its apparent public health importance, myxin B, 2,500 U; trimethoprim, 5 mg; and vancomy- however, the natural reservoirs and sources of cin, 10 mg (3). Plates were incubated for 48 h at 42°C Campylobacter for human infections have not in an atmosphere of 5% oxygen, 10% carbon dioxide, yet been fully determined. and 85% nitrogen. Typical flat gray watery nonhemo- Among domestic animals, C. fetus subsp. je- lytic spreading colonies were Gram stained (counter- juni is known to cause in sheep (17), stained with a 0.8% solution of carbol fuchsin), and winter scours in cattle (9), infectious in those colonies consisting of small gram-negative chickens (17), and possibly enteritis in dogs curved or spiral rods were further isolated. Identifica- (2). tion of C. fetus subsp. jejuni was based on typical In wild animal populations, it has been isolated vibrio forms, colony characteristics, oxidase and cata- from bank voles (6), pigeons, blackbirds, star- lase positivity, darting motility on dark-field exami- lings, and sparrows (16). The objective of the nation, lack of fermentation or oxidation of glucose, present study was to determine the frequency of production of hydrogen sulfide as detected by lead C. fetus subsp. jejuni in migratory waterfowl. acetate paper over Kligler slants, growth at 42°C but The observed carrier rate of 35% may make not at 25°C, and sensitivity to nalidixic acid (30-g these waterfowl a potential health hazard for disks) (1). Isolates were cultured only under humans. microaerophilic conditions, and no attempt was made was in at the 80th to test growth in other atmospheres such as 21% (This paper presented part oxygen. Annual Meeting of the American Society for Statistical comparisons were made with the chi- Microbiology, Miami Beach, Fla. [Abstr. Annu. square test. Paired data for cecal versus cloacal spec- Meet. Am. Soc. Microbiol. 1980, Q76, p. 206].) imens were analyzed with the McNemar test for cor- MATERIALS AND METHODS related proportions (14). From November 1978 through January 1980, 445 RESULTS wild ducks were collected by cooperating hunters in 154 northern Colorado. The waterfowl were cultured im- Of the 445 migratory waterfowl studied, mediately in the field or were transported to the (35%) carried C. fetus subsp. jejuni. The fre- laboratory within 24 h at subfreezing temperatures. quency of isolation among different species is Ceca were opened under sterile conditions and samples shown in Table 1. Shovelers had significantly of cecal contents were collected with sterile swabs. In higher (P < 0.001) and green-winged teal had 64 of the ducks, cloacal swabs were obtained along significantly lower (P < 0.01) isolation rates than with cecal specimens to compare the isolation rates the remaining ducks. Gadwalls had a low isola- t Present address: Enteric Diseases Branch, Bureau of tion rate (15%), but the numbers were too small Epidemiology, Center for Disease Control, Atlanta, GA 30333. to achieve statistical significance. The 17 ducks 406 VOL. 12, 1980 C. FETUS SUBSP. JEJUNI FROM WATERFOWL 407 TABLE 1. Isolation of C. fetus subsp. jejuni from (15%) and green-winged teal (16%), which feed cecal contents of various species of migratory almost exclusively on vegetable matter. In con- waterfowl trast, shoveler ducks, which had a 66% preva- lence of C. fetus subsp. often strain mud No. jejuni, Species of waterfowl tested posi-No. % from the bottom of ponds and rivers to extract tive molluscs and other animal foods that constitute Shoveler (Spatula cly- 35 23 66 approximately one-third of their diet (11). The peata) animal food intake, as weil as the isolation rates Pintail (Anas acuta) 30 15 50 of C. fetus subsp. jejuni, of the other ducks American widgeon (Mer- 38 16 42 ranged between those of green-winged teal and eca americana) shovelers. Hussong et al. (8) have shown that Mallard (Anas platyrhyn- 243 82 34 chos) different species of waterfowl and also waterfowl Green-winged teal (Anas 56 9 16 of the same species maintained on different diets carolinensis) harbor substantially different total numbers of Gadwall (Anas strepera) 26 4 15 intestinal . It is, therefore, possible that Other species 17 5 29 dietary variation among the species studied Totals 445 154 35 could account for some of the differences in isolation rates of C. fetus subsp. jejuni. of "other" species collected in numbers too smail The question was raised during the course of for separate statistical analysis included redhead the study whether live birds could be trapped, (Aythya americana), bufflehead (Bucephala al- cloacal swabs obtained, and the birds subse- beola), and lesser scaup (Aythya affinis). Carrier quently released unharmed. However, the iso- rates of hens compared with drakes were similar lation rates of C. fetus subsp. jejuni were so for all species, with the exception of mallard much higher from cecal specimens than from hens which had a significantly higher carrier cloacal swabs that cecal samples were used rate of C. fetus subsp. jejuni than did mallard throughout the study. Whether higher isolation drakes (50 versus 31%, P < 0.05). rates from other species of animals would simi- There was no significant difference in carrier larly be obtained by using cecal specimens in- rates between resident Colorado ducks collected stead of cloacal or rectal swabs is presently in October (33%) and migrating northern ducks unknown. collected later in the season (35%). Carrier rates Although C. fetus subsp. jejuni has been in- for all species were similar from one year to the criminated in avian hepatitis (a contagious dis- next. ease of chickens characterized by diarrhea, A comparison of culture techniques in 64 weight loss, liver lesions, and drop in egg pro- ducks showed that C. fetus subsp. jejuni was duction) (7), none of the waterfowl in this study isolated from both cecal and cloacal swabs in 9 showed gross lesions of the liver or gut, and all ducks, from only cecal swabs in 11 ducks, and were apparently healthy when collected. The from only the cloacal swab in 1 duck. The higher high body temperature (42°C) of waterfowl and rate of isolation with cecal swabs was statisti- other birds may encourage carriage of this or- cally significant (P < 0.001). ganism, whose optimum temperature for growth is also 42°C. DISCUSSION The sources of C. fetus subsp. jejuni for hu- The north central plateau of Colorado east of man infections remain to be fully elucidated. the Rocky Mountains is part of the Central have been isolated from fresh- Flyway for migratory birds. Its many rivers, water lakes and streams (10), and one large streams, warm water sloughs, ponds, shallow waterborne outbreak of Campylobacter infec- reservoirs, and abundant fields of grain attract tion occurred without a known source (18). Pre- hundreds of thousands of migratory ducks vious data from this laboratory showed that yearly, with estimated peak daily counts reach- Campylobacter survives in stream water for over ing over 200,000 in the Platteville area during 1 week at 4°C (3). Waterfowl have been reported the 1978 and 1979 hunting seasons (Colorado to occasionally harbor other enteric pathogens State Fish and Game Department, personal such as Salmonella (5, 12), and heavy concen- communication). This study showed that 35% of trations of wild waterfowl can cause a significant a sample of 445 of these migratory waterfowl elevation in fecal coliform counts of surface wa- carried C. fetus subsp. jejuni. ter and sediment (8). This study raises the pos- The species variation noted in isolation rates sibility that migratory waterfowl may be yet may be related to differences in feeding habits. another reservoir for C. fetus subsp. jejuni. The lowest carriage rates were found in gadwalls Whether they contaminate freshwater ponds 408 LUECHTEFELD ET AL. J. CLIN. MICROBIOL. and streams with this enteric pathogen or ac- Microbial impact of Canada geese (Branta canadensis) and whistling swans (Cygnus columbianus columbi- quire the infection from water already polluted anus) on aquatic ecosystems. Apple. Environ. Microbiol. by the feces of domestic animals is not presently 37:14-20. known. 9. Jones, F. S., and R. B. Little. 1931. Vibrionic enteritis in calves. J. Exp. Med. 53:845-851. LITERATURE CITED 10. Knill, M., W. G. Suckling, and A. D. Pearson. 1978. 1. Blaser, M. J., I. D. Berkowitz, F. M. LaForce, J. Environmental isolation of heat-tolerant Campylobac- Cravens, L. B. Reller, and W.-L. L. Wang. 1979. ter in the Southampton area. Lancet ii:1002-1003. Campylobacter enteritis: Clinical and epidemiological 11. Kortright, F. H. 1943. The ducks, geese, and swans of features. Ann. Intern. Med. 91:179-185. North America. American Wildlife Institute, Washing- 2. Blaser, M. J., J. Cravens, B. W. Powers, and W.-L. L. ton, D.C. Wang. 1978. Campylobacter enteritis associated with 12. Mitchell, T. R., and T. Ridgwell. 1971. The frequency canine infection. Lancet ii:979-981. of salmonella in wild ducks. J. Med. Microbiol. 4:359- 3. Blaser, M. J., H. L. Hardesty, B. Powers, and W.-L. 361. L. Wang. 1980. Survival of Campylobacter fetus subsp. 13. Pai, C. H., S. Sorger, L. Lackman, R. E. Sinai, and M. jejuni in biological milieus. J. Clin. Microbiol. 11:309- I. Marks. 1979. Campylobacter in chil- 313. dren. J. Pediatr. 94:589-591. 4. Bruce, D., W. Zochowski, and I. R. Ferguson. 1977. 14. Remington, R. D., and M. A. Schork. 1970. Statistics Campylobacter enteritis. Br. Med. J. 2:1219. with applications to biological and health sciences. 5. Faddoul, G. P., and G. W. Fellows. 1966. A five-year Prentice-Hall, Inc., Englewood Cliffs, N.J. survey of the incidence of salmonella in avian species. 15. Skirrow, M. B. 1977. Campylobacter enteritis: a "new" Avian Dis. 10:296-304. disease. Br. Med. J. 2:9-11. 6. Fernie, D. S., and R. W. A. Park. 1977. The isolation 16. Smibert, R. M. 1969. Vibrio fetus var. intestinalis isolated and nature of Campylobacters (microaerophilic vibrios) from the intestinal contents of birds. Am. J. Vet. Res. from laboratory and wild rodents. J. Gen. Microbiol. 30:1437-1442. 10:325-329. 17. Smibert, R. M. 1978. The genus Campylobacter. Annu. 7. Hagan, J. R. 1964. Diagnostic techniques in avian vib- Rev. Microbiol. 32:673-709. rionic hepatitis. Avian Dis. 8:428-437. 18. Tiehan, W., and R. L. Vogt. 1978. Waterborne campyl- 8. Hussong, D., J. M. Damere, R. J. Limpert, W. J. L. obacter gastroenteritis-Vermont. Morbid. Mortal. Sladen, R. M. Weiner, and R. R. Colwell. 1979. Weekly Rep. 27:207.