Abstract
Antibiotic resistance of Escherichia coli from sows and pigs was determined to compare patterns between pigs of various ages and degrees of antibiotic use. Resistance patterns differed between farm types and pigs of differing ages, indicating that pig age and degree of antibiotic use affect resistance of fecal E. coli.
Antibiotics are used in livestock systems to combat disease and improve animal productivity. Feed-based antibiotics consistently benefit productivity, increasing the ability of farms to maintain profitable margins. Benefits of antibiotic use include improved growth (7, 20), decreased nitrogen excretion and thus reduced environmental impact (17), and decreased pathogen loads (11).
Contrasting the above benefits are suggestions that agricultural use of antibiotics may be partly responsible for the emergence of drug-resistant organisms (4). While a number of studies have been conducted and conferences have been assembled to address this issue (3, 8), a lack of data continues to hamper efforts to devise solutions. Information on resistance prevalence and effect of farm management is especially lacking. Much of the current data are derived from clinical isolates, and as such, these data may be biased by age and condition of the animals and level of antibiotic use. So that more definitive recommendations can be devised, it is important that confounding factors be characterized. This study was designed to determine effects of pig age and level of antibiotic use on single and multiple antibiotic resistance patterns.
Ten swine farms in various regions of Tennessee were selected for this study. All farms were typical of U.S. production, with sizes ranging from approximately 1,500 to 10,000 pigs produced annually. Farms were classified as low antibiotic use (LU) (n = 3) if subtherapeutic feed-based antibiotics were not used or if only subtherapeutic concentrations of tetracyclines were used for brief periods. Farms classified as high antibiotic use (HU) (n = 7) routinely used subtherapeutic feed-based antibiotics and/or injectable antibiotics.
Five sows were randomly selected from each farm, and fecal material was collected via rectal swab at 7 days postpartum. Swabs were also obtained from five pigs from each test sow. Pigs were identified by ear notches so that additional samples could be collected at 35 and 63 days of age. Escherichia coli was isolated by culture on lactose MacConkey agar prior to confirmation by biochemical analysis (API20; Vitek bioMerieux, Syosset, N.Y.). Isolates were subjected to standardized disk diffusion tests (5, 16) against apramycin (15 μg), carbadox (20 μg), gentamicin (10 μg), neomycin (30 μg), and oxytetracycline (30 μg). Isolates were determined to be resistant if zone sizes were less than 14 mm for apramycin, 11 mm for carbadox, 14 mm for gentamicin, 16 mm for neomycin, and 18 mm for oxytetracycline.
Data on resistance to individual drugs were summarized by establishment of multiple-drug resistance groups and analyzed in two- and three-way contingency tables to test differences in frequencies of multiple-drug resistance groups across farms, pigs or sows, and time. Fisher’s exact test and Cochran-Mantel-Haenszel tests (1) were used where appropriate, using SAS software (18).
Table 1 provides a summary of confirmed use of test antibiotics for each farm group. While farms were separated into LU and HU categories based on degree of antibiotic use, we are unable to provide a more detailed history of use for each antibiotic since some producers on HU farms could not reliably provide detailed drug use history beyond a few months prior to the study. We thus felt that specific indications on a farm-by-farm basis were less reliable and might result in erroneous conclusions.
TABLE 1.
Use of test antibiotics within 1 year previous to survey
| Antibiotic | Method of presentation | Confirmed usea on:
|
|
|---|---|---|---|
| HU farms | LU farms | ||
| Apramycin | Feed (sulfate form, 165 mg/kg of feed) | × | |
| Carbadox | Feed (55 mg/kg of feed) | × | |
| Gentamicin | Water (sulfate form, 6.6 mg/liter of drinking water) | × | |
| Neomycin | Water (sulfate form, 22 mg/kg of body weight) | × | |
| Oxytetracycline | Feed (110 mg/kg of feed) | × | × |
| Oxytetracycline | Injected (HCl form, 10 mg/kg of body weight) | × | |
Confirmed use on one or more farms within the group.
On LU farms the incidence of multiple resistance generally remained constant throughout the various growth phases. On HU farms, incidence of multiple resistance was greatest (P < 0.001) in E. coli from pigs at 35 days of age (Table 2), except for patterns including neomycin and oxytetracycline, which increased at 63 days of age. In general, the incidence of resistance was lower at all pig ages on LU farms than on HU farms. No E. coli isolates from LU farms showed single resistance to apramycin (A), and only a few such isolates were noted from HU farms. Instead, the majority of apramycin resistance occurred as part of a multiple resistance pattern. Similarly, few isolates were found to be resistant only to neomycin (N); instead, neomycin resistance generally occurred as part of a multiple resistance pattern. In contrast, oxytetracycline (O) resistance was often found as a single resistance. The greatest apramycin resistance was found in 35-day-old pigs on all seven HU farms, with a similar pattern observed for gentamicin on six of the seven HU farms (data not shown). In contrast, apramycin and gentamicin resistance did not increase at that pig age on any of the 3 LU farms. On LU farms, the most common combinations included C-G-N-O, C-G-O, and G-N-O (Table 3). These same patterns were found on HU farms, and in most cases the incidence was greater than that on LU farms. We observed a marked increase in the incidence of A-C-G-N-O, A-C-G-O, and A-G-N-O multiple resistance patterns in E. coli from HU farms in pigs at 35 days of age.
TABLE 2.
Incidence of single and multiple resistance to tested antibiotics in fecal E. coli from sows and pigs of various ages exposed to high and low antibiotic usage
| Drug and resistance pattern | % of isolates with indicated resistance pattern from animals and farm type (n)a
|
|||||||
|---|---|---|---|---|---|---|---|---|
| Sows
|
Pigs, 7 days
|
Pigs, 35 days
|
Pigs, 63 days
|
|||||
| HU (322) | LU (140) | HU (1,731) | LU (747) | HU (1,683) | LU (729) | HU (999) | LU (407) | |
| Apramycin | ||||||||
| Single | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 |
| Multiple | 0.3 | 0.7 | 8.2 | 9.2 | 60.9 | 7.1 | 29.5 | 5.8 |
| No resistance | 99.7 | 99.3 | 91.8 | 90.8 | 39.1 | 92.9 | 70.5 | 94.2 |
| Carbadox | ||||||||
| Single | 0 | 0.7 | 0.1 | 0.9 | 1.7 | 0 | 0 | 0.2 |
| Multiple | 17.7 | 37.8 | 35 | 17.8 | 55.3 | 17.1 | 47.3 | 20.6 |
| No resistance | 82.3 | 61.5 | 64.9 | 81.3 | 43.0 | 82.9 | 52.7 | 79.2 |
| Gentamicin | ||||||||
| Single | 0 | 3.6 | 0.5 | 5.9 | 0.1 | 6.4 | 0 | 0 |
| Multiple | 74.8 | 82.8 | 73.3 | 75.8 | 91.9 | 71.0 | 85.9 | 72.5 |
| No resistance | 25.2 | 13.6 | 26.2 | 18.3 | 8.0 | 22.6 | 14.1 | 27.5 |
| Neomycin | ||||||||
| Single | 0 | 0 | 0.2 | 0 | 0 | 0 | 0 | 0 |
| Multiple | 56.2 | 53.6 | 55.2 | 44.6 | 54.8 | 41.7 | 64.6 | 34.3 |
| No resistance | 43.8 | 46.4 | 44.6 | 55.4 | 45.2 | 58.3 | 35.4 | 65.7 |
| Tetracycline | ||||||||
| Single | 17.4 | 5.0 | 16.7 | 10.8 | 2.9 | 11.0 | 5.2 | 22.6 |
| Multiple | 81.7 | 79.3 | 79.5 | 71.8 | 91.4 | 78.9 | 93.6 | 76.9 |
| No resistance | 0.9 | 15.7 | 3.8 | 17.4 | 5.7 | 10.1 | 1.2 | 0.5 |
The approximate least significant difference between percentages ranges from 3 to 8.
TABLE 3.
Incidence of single and multiple resistance patterns in fecal E. coli from sows and pigs exposed to high and low antibiotic use
| Resistance pattern | % of isolates with indicated resistance pattern from animals and farm type (n)a
|
|||||||
|---|---|---|---|---|---|---|---|---|
| Sows
|
Pigs, 7 days
|
Pigs, 35 days
|
Pigs, 63 days
|
|||||
| HU (322) | LU (140) | HU (1,731) | LU (747) | HU (1,683) | LU (729) | HU (999) | LU (407) | |
| A-C-G | 0 | 0 | 0 | 0 | 1.5 | 0 | 0 | 0 |
| A-C-G-N | 0 | 0 | 0 | 0 | 1.2 | 0 | 0 | 0 |
| A-C-G-N-O | 0.3 | 0 | 1.3 | 0.1 | 19.7 | 0.1 | 14.8 | 2.0 |
| A-C-G-O | 0 | 0 | 0.6 | 0.1 | 19.2 | 0.3 | 3.2 | 2.7 |
| A-C-N-O | 0 | 0.7 | 0 | 0 | 0.1 | 0 | 0 | 0 |
| A-C-O | 0 | 0 | 0 | 0 | 0.1 | 0 | 0 | 0 |
| A-G | 0 | 0 | 0 | 0.1 | 0.1 | 0 | 0 | 0 |
| A-G-N | 0 | 0 | 0 | 1.5 | 0 | 0 | 0 | 0 |
| A-G-N-O | 0 | 0.7 | 5.4 | 6.6 | 13.0 | 0.7 | 7.6 | 0.3 |
| A-G-O | 0 | 0 | 0.9 | 0.8 | 6 | 5.9 | 3.7 | 1.0 |
| A-N-O | 0 | 0 | 0 | 0 | 0 | 0 | 0.1 | 0 |
| A-O | 0 | 0 | 0 | 0.9 | 0.1 | 0.1 | 0.1 | 0 |
| C-G | 0 | 0.7 | 0.1 | 0 | 0.4 | 0 | 0.3 | 0 |
| C-G-N | 0 | 0.7 | 0.1 | 2.1 | 0.3 | 0.1 | 0.3 | 0 |
| C-G-N-O | 10.6 | 20.7 | 0.1 | 7.4 | 6.1 | 5.4 | 13.8 | 5.9 |
| C-G-O | 3.8 | 14.3 | 7.9 | 6.3 | 4.3 | 4.0 | 10.5 | 7.4 |
| C-N | 0 | 0 | 3.4 | 0 | 0 | 0 | 0 | 0 |
| C-N-O | 0.9 | 0 | 0.1 | 0.13 | 0.6 | 0.4 | 1.9 | 0.3 |
| C-O | 2.2 | 1.4 | 0.4 | 1.6 | 1.8 | 6.9 | 2.5 | 2.5 |
| G-N | 0.3 | 5.0 | 1.0 | 3.5 | 0.2 | 2.2 | 0.1 | 0.3 |
| G-N-O | 40.1 | 25.0 | 33.1 | 21.8 | 13.0 | 30.0 | 22.1 | 23.8 |
| G-O | 19.9 | 15.7 | 19.5 | 25.4 | 6.8 | 22.4 | 9.4 | 29.2 |
| N-O | 4.0 | 1.4 | 5.8 | 1.1 | 0.7 | 2.7 | 3.8 | 2.0 |
| A | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 |
| C | 0.1 | 0.7 | 0.1 | 0.9 | 1.7 | 0 | 0 | 0.3 |
| G | 0 | 3.6 | 0.6 | 5.9 | 0.1 | 6.5 | 0 | 0 |
| N | 0 | 0 | 0.2 | 0 | 0 | 0 | 0 | 0 |
| O | 17.4 | 5.0 | 16.7 | 10.8 | 2.9 | 11.0 | 5.2 | 22.6 |
| Sensitive to all | 0.6 | 5.0 | 1.6 | 3.35 | 0.1 | 1.4 | 0.5 | 0 |
Standard errors were approximately 10% of the percentages, so least significant differences ranged from approximately 1 to 10, with higher values occurring with the larger percentages.
Whereas survey data on resistance are commonly reported, age of the animals is generally not taken into account. Our data indicate that age is an important factor affecting resistance in swine. The increased incidence of resistance noted in nursery pigs may be a reflection of increased antibiotic use at that time, and may also reflect the increased colonization by pathogens that occurs postweaning (14, 15), in which resistance may occur more commonly than in commensal organisms.
Cross-resistance within the aminoglycoside group was common. One of the most common combinations was N-G-O. The two aminoglycosides in this pattern, neomycin and gentamicin, share more commonalities in structure and function than with apramycin. E. coli from pigs had a greater incidence of resistance to A-G-N-O, while E. coli from sows had greater resistance to C-G-N-O. Apramycin was a common feed additive reportedly used on HU farms in nursery-age pigs; these data may indicate that transfer of organisms occurs between nurseries and farrowing (birthing) rooms. If that is the case, the contamination appears to have a greater effect in pigs than in sows. This may be due to resistant bacteria that specifically colonize the young pig or a greater resistance by the sow to invasion by bacteria from an outside source. This would support other reports suggesting that pigs, and not sows, are the primary sources of E. coli in farrowing barns (2, 6). Hinton and Linton (9) and Katouli et al. (10) determined that young pigs are able to maintain a unique microfloral population even while in close contact with the sow. Other investigators also observed higher percentages of resistant bacteria in young mammals than in adults (12, 13, 19, 22). Because younger swine are at a greater risk of contracting disease (21), they likely represent the most common source of clinical isolates, which in turn are more likely to be the source of resistance. It is evident from this study that resistance data from young pigs should not be extrapolated for estimates in market-age swine, which have the greatest relevance to pork products. Further studies to determine other factors affecting resistant bacteria in livestock systems appear warranted.
Acknowledgments
This work was supported in part by USDA HATCH funds allocated to the University of Tennessee Agricultural Experiment Station.
We thank all swine managers and producers that cooperated in this study.
REFERENCES
- 1.Agresti A. Categorical data analysis. New York, N.Y: John Wiley Inc.; 1990. [Google Scholar]
- 2.Arbuckle J B R. The distribution of certain Escherichia coli strains in pigs and their environment. Br Vet J. 1968;124:152–159. doi: 10.1016/s0007-1935(17)39454-x. [DOI] [PubMed] [Google Scholar]
- 3.ASM Task Force on Antibiotic Resistance. Report of the ASM Task Force on Antibiotic Resistance. Washington, D.C: American Society for Microbiology; 1995. [Google Scholar]
- 4.Bates J. Epidemiology of vancomycin-resistant enterococci in the community and the relevance of farm animals to human infection. J Hosp Infect. 1997;37:89–101. doi: 10.1016/s0195-6701(97)90179-1. [DOI] [PubMed] [Google Scholar]
- 5.Bauer A W, Kirby M M, Sherris J L, Turck M. Antibiotic susceptibility testing by a standardized single disk method. Am J Clin Pathol. 1966;45:493–496. [PubMed] [Google Scholar]
- 6.Callear J F, Smith I M. The effect of furazolidone on Escherichia coli infection of newborn piglets. Br Vet J. 1966;122:169–179. [Google Scholar]
- 7.Cromwell G L, Davis G W, Morrow W E, Primo R A, Borzeboom D W, Sims M D, Staisiewske E P, Ho C H. Efficacy of the antimicrobial compound U-82, 127 as a growth promoter for growing-finishing pigs. J Anim Sci. 1996;74:1284–1287. doi: 10.2527/1996.7461284x. [DOI] [PubMed] [Google Scholar]
- 8.Gustafson R H. Use of antibiotics and human health concerns. J Dairy Sci. 1991;74:1428–1432. doi: 10.3168/jds.S0022-0302(91)78299-4. [DOI] [PubMed] [Google Scholar]
- 9.Hinton M, Linton A H. The ecology of Escherichia coli in healthy newborn piglets. Br Vet J. 1987;143:541–548. doi: 10.1016/0007-1935(87)90044-3. [DOI] [PubMed] [Google Scholar]
- 10.Katouli M, Lund A, Wallgren P, Kuhn I, Söderlind O, Möllby R. Phenotypic characterization of intestinal Escherichia coli of pigs during suckling, postweaning, and fattening periods. Appl Environ Microbiol. 1995;61:778–783. doi: 10.1128/aem.61.2.778-783.1995. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Kyriakis S C, Sarris K, Kritas S K, Tsinas A C, Giannakopoulos C. Effect of salinomycin in the control of Clostridium perfringens type C infections in suckling pigs. Vet Rec. 1996;138:281–283. doi: 10.1136/vr.138.12.281. [DOI] [PubMed] [Google Scholar]
- 12.Langlois B E, Dawson K A, Cromwell G L, Stahly T S. Antibiotic resistance in pigs following a 13 year ban. J Anim Sci. 1986;62:18–32. [Google Scholar]
- 13.Linton K B, Lee P A, Richmond M H, Gillespie W A. Antibiotic resistance and transmissible R-factors in the intestinal coliform flora of healthy adults and children in an urban and a rural community. J Hyg Camb. 1972;70:99–104. doi: 10.1017/s0022172400022130. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Mathew A G, Sutton A L, Scheidt A B, Patterson J A, Kelly D T, Meyerholtz K A. Effect of galactan on selected microbial populations and pH and volatile fatty acids in the ileum of the weanling pig. J Anim Sci. 1993;71:1503–1509. doi: 10.2527/1993.7161503x. [DOI] [PubMed] [Google Scholar]
- 15.McAllister J S, Kurtz H J, Short E C. Changes in the intestinal flora of young pigs with postweaning diarrhea or edema disease. J Anim Sci. 1979;49:868–879. doi: 10.2527/jas1979.493868x. [DOI] [PubMed] [Google Scholar]
- 16.National Committee for Clinical Laboratory Standards. Performance standards for antimicrobic disk susceptibility tests. 4th ed. Villanova, Pa: National Committee for Clinical Laboratory Standards; 1990. [Google Scholar]
- 17.Roth F X, Kirchgessner M. Influence of avilamycin and tylosin on retention and excretion of nitrogen in finishing pigs. J Anim Physiol Anim Nutr. 1993;69:245–250. [Google Scholar]
- 18.SAS Institute. SAS/STAT software: changes and enhancements through release 6.12. Cary, N.C: SAS Institute; 1997. [Google Scholar]
- 19.Sogaard H. Incidence of drug resistance and transmissible R factors in strains of E. coli from faeces of healthy pigs. Acta Vet Scand. 1973;14:381–391. doi: 10.1186/BF03547426. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Stahly T S, Cromwell G L, Monegue H J. Effects of dietary inclusion of copper and(or) antibiotics on the performance of weanling pigs. J Anim Sci. 1980;51:1347–1351. doi: 10.2527/jas1981.5161347x. [DOI] [PubMed] [Google Scholar]
- 21.United States Department of Agriculture. National Swine Survey. Washington, D.C: National Animal Health Monitoring System (NAHMS), U.S. Department of Agriculture; 1992. Morbidity/mortality and health management of swine in the United States; pp. 38–42. [Google Scholar]
- 22.Wierup M. Antibiotic resistance and transferable antibiotic resistance of Escherichia coli isolated from Swedish calves 5 and 30 days old. Nord Vetmed. 1975;27:77–84. [PubMed] [Google Scholar]