| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
ANIMAL PRODUCTION |
* Department of Animal Science, Michigan State University, East Lansing 48854;
and
Department of Animal Sciences, University of Illinois at Urbana-Champaign, Urbana 61801; and
and
Alltech, Inc., Dunboyne, Ireland
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionImplicationsLiterature Cited |
|---|
Key Words: Antimicrobial • Mannan Oligosaccharide • Swine
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionImplicationsLiterature Cited |
|---|
; Stewart et al., 2001). Type I fimbriae specifically bind to glycoproteins that contain mannose on the intestinal cell surface (Ofek et al., 1977). The cell walls of yeasts such as Saccharomyces cerevisiae contain mannan components, which can be isolated industrially to produce feed additives known as mannan oligosaccharides (MOS; Spring et al., 2000). Bio-Mos (Alltech, Inc., Nicholasville, KY) is such a product that has been shown to bind, in vitro, to bacterial cells possessing Type 1 fimbriae, including species of Escherichia coli and Salmonella (Newman, 1994; Spring et al., 2000), preventing these pathogens from binding to and proliferating at the mucosal surface of the intestine. Bio-Mos also has been shown to alter immune function (Savage and Zakrzewska, 1996; Davis et al., 2004).
Modest increases in growth rates of weanling pigs have been supported when MOS is added to the diet at levels of 1 to 4 g/kg (Miguel et al., 2003). If the perceived action of MOS is as just described, the benefit of the product may be greater in the presence of greater disease challenges. In fact, there is support for a greater response to MOS in situations in which growth rate is slower (Miguel et al., 2003). Similarly, antimicrobials enhance growth and decrease mortality in young pigs, with even greater response under high-disease, stressful conditions (Cromwell, 2001). The potential role of MOS in improving the health and performance of pigs is of increasing interest because of public concern about the use of antimicrobials in pig production.
The objective of this experiment was to evaluate the effects of MOS in the presence or absence of an antimicrobial product on the growth performance and indices of health of weaned pigs and to determine whether these effects are greater on farms with greater challenges to pig health.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionImplicationsLiterature Cited |
|---|
The experimental protocol used in this study was approved by the All-University Committee on Animal Use and Care at Michigan State University (AUF No. 006/99–086–00).
Farms
Pigs were reared in nursery facilities on three different farms, two of which were commercial farms (A and B) and one was located at Michigan State University (C). The two commercial farms were selected based on the hospitality of farm owners/managers and on the nursery pig growth performance observed by Michigan State University researchers in a previous study that was conducted <12 mo earlier. Average daily gain during that 7-wk growth study was 0.39 and 0.57 kg/d for Farms A and B, respectively. In comparison, ADG by pigs at Farm C during the same time was 0.45 kg/d. The variation in growth performance among the three farms was intentionally sought as a possible, though not clinically proven, indicator of the varied health statuses on these farms. No major structural or operational changes had been made to the nurseries on these farms since the time that the previous research was conducted.
The nursery building on Farm A housed a total of 4,000 pigs. It was a converted 1,000-animal, mechanically ventilated, finisher building. Portions of the facility, including the 1.95-m deep pit, all exterior walls, and the roof, were >15 yr old. In this building, eight rooms shared a common aisle that ran the entire length of the building down the middle of each room. Solid wood walls and doors in the aisle separated rooms above the flooring, and a concrete wall separated the manure pits of rooms beneath the flooring. Each room was mechanically ventilated, with air inlets at ceiling height in the west wall and two variable-speed fans mounted in the east wall. At floor level, there were 12 pens per room, each containing 32 or 33 pigs. Two rooms, providing a total of 24 pens, were used in this experiment. Pens were 1.83 m x 5.49 m with solid, plastic side panels mounted to the exterior wall and vertical, steel rod gates along the center aisle. Flooring was made of round steel rods. A heated, 50-cm-wide, concrete pad, parallel to the aisle and approximately 1.5 m from the outside wall, ran continuously through all pens on each side of the aisle. Two nipple-type waterers and a round feeder (AP Hog Diner; Automated Production Systems, Assumption, IL) were provided per pen. All-in/all-out management was practiced by room, with two rooms emptied, cleaned, dried, and populated each week. Personnel and pig traffic between rooms was necessary; thus, the entire nursery was managed on a continuous flow biosecurity basis. In addition to the floor pens that were used in the present study, each room also had eight small pens decked above floor pens for approximately 10 small and weak pigs per pen. These pens were not used in the present study and were occupied by nonresearch pigs.
Farm B housed 6,400 pigs and was <5 yr old at the time of our study. The nursery had eight identical rooms, each with three doorways opening into a common hallway. There were 40 pens per room (1.68 m x 3.05 m) with an aisle along all walls and an aisle down the center of the room (20 pens on either side). The experiment was conducted with 32 pens or 16 pen pairs; each pair shared a single feeder (double-sided, six-hole, stainless steel) and was identified as a single experimental unit. Each pen contained 17 to 19 pigs. The hallway was on the east side of the building and ran the entire length of the building. Each room was mechanically ventilated, with air inlets in the east wall (hallway) and exhaust fans in the west wall. Steel flooring (TriBar; Nooyen Flooring Inc., Muncie, IN) was used throughout each room over a pit that was 1.22 m deep. All fencing and gates were horizontal steel bar and rod. There were two nipple waterers per pen. The nursery was operated all-in/all-out by room, and each week one room was emptied, cleaned, refilled, and had its shallow pit drained and recharged with fresh water.
Farm C was the Michigan State University Swine Farm, with the nursery part of a 250-sow farrow-to-finish confinement complex built from 1996 to 1998. Sixteen different production rooms were connected by a common hallway, with doors that separated sections or groups of rooms designated for the breeding, gestation, farrowing, nursery, and grow-finish phases of production. The nursery section consisted of four rooms, each housing up to 240 pigs. Nursery rooms were mechanically ventilated, with fans in the north wall pulling preheated air from the hallway through the room. There were 30 pens per room, each 1.22 m x 1.83 m, with only 24 pens and four pigs per pen used in this experiment. Each nursery room had one door leading to the hallway and an emergency door leading to the outside. An aisle surrounded a center island of 16 pens (eight per row, with pens back-to-back). There were eight additional pens along both the east and west walls. Round-rod steel flooring over a 1.22 m deep pit was used in all pens. All aisles were concrete. Fences and gates were made of vertical fiberglass rod. One stainless steel single-sided, two-hole feeder was provided in each pen. Each pen also had one cup waterer (LaBuvette; Gameco Pty. Ltd., Brisbane, Australia). The entire complex, including nursery section, was operated all-in/all-out by room, with one nursery room emptied, cleaned, and refilled every 2 wk.
Animals
Genotype of pigs varied across farms. Farms A, B, and C pigs were (Yorkshire x Landrace) x T-Max (Premier Swine Breeding Systems, LLC, Michigantown, IN), GIS (Genetic Improvement Services, Newton Grove, NC) x PIC (Pig Improvement Company, Franklin, KY), and (Yorkshire x Landrace) x Duroc, respectively. Farm B did not identify offspring as coming from specific sire and dam lines. Pigs on Farms A and B were commingled from two sow farms. A total of 1,648 pigs was identified one day before weaning for use in this 2 x 2 factorial experiment. Number of pigs allotted (n = 1,443), weaning age, and initial weight are provided in Table 1. All animals were moved into their respective nurseries on a single day.
|
The two factors in the 2 x 2 factorial design were 1) MOS (0 or 0.3% in Phase I and 0.2% in Phases II, III, and IV; Bio-Mos, Alltech, Inc.); and 2) antimicrobial (AM; 0 or 110 mg of tylosin and 110 mg of sulfamethazine/kg of diet in all phases; Tylan 40 Sulfa-G; Elanco Animal Health., Indianapolis, IN). All diets contained Cu and Zn at or slightly above requirement levels estimated by NRC (1998). These treatments were imposed over four nursery phases (I, II, III, and IV; 4, 7, 14, and 17 d, respectively). Diet formulations are shown in Table 2. Phase I and II diets were in 3-mm diameter pellets, and Phase III and IV diets were mash. All diets were commercially manufactured (I and II at United Feeds, Sheridan, IN and III and IV at Hubbard Feeds, Shipshewana, IN), portioned in 22.7-kg bags, and transported to the three farms in semitrailers that were parked and left at the farm until the study was completed. Feed was removed from the trailers when needed. Bags of feed were labeled such that farm employees did not know the dietary treatments.
|
Pigs were weighed individually at the beginning of the experiment, at the end of Phase II (d 11), and at the end of Phase IV (d 42). Feed disappearance was recorded by counting the number of bags added to each feeder. Feed remaining in each feeder at the end of each phase was removed and weighed. Data measured included ADG, ADFI, and G:F.
Health
The health status of the pigs differed among farms and was in part determined by a written survey completed by the respective owners and managers, but it was not verified by veterinary consultation or diagnostics immediately preceding this study. On Farm A, pigs were described as free of porcine reproductive and respiratory syndrome (PRRS), but slightly infected with Actinobacillus pleuropneumoniae (App), Mycoplasma hyopneumoniae (MPS), atrophic rhinitis, and slight diarrhea caused by Escherichia coli, rotaviruses, and Salmonella spp. Farm A also experienced moderate infections and some death loss caused by Streptococcus suis and Staphylococcus spp. Pigs on Farm B were infected slightly with PRRS, APP, and MPS, with no disease considered moderate, serious, or a cause of significant number of deaths. Pigs on Farm C were PRRS and APP free, but were known to have slight infections of Streptococcus suis and Lawsonia intracellularis, which led to the euthanasia of approximately 0.5% of weanling pigs.
When researchers from Michigan State University conducted a previous study on these farms (unpublished), average rates of combined removal and mortality observed in each nursery were 0.3, 12.3, and 1.5 for Farms A, B, and C, respectively. The term "removal" was used in the present study and also may be clinically referred to elsewhere as "morbidity." Removal represented piglets that were moved from their pen of origin to a hospital pen, where they were provided with more space, supplemental heat, observation, possible pharmaceutical treatment, and different diet.
Each farm had a different set of standard operating procedures pertaining to health care during our study. Farm A treated sick pigs with injectable tylosin and ceftiofur sodium. Lameness, labored breathing, and loose stools were the major symptoms treated. Farm B did not treat sick pigs with injectable antimicrobials, but removed them to a hospital pen. Farm C used penicillin as an injectable treatment for sick pigs. Farm A provided electrolytes (BlueLite; TechMix, Inc., Stewart, MN) and chlortetracycline/sulfamethazine in the water to all pigs from d 12 to 42 of the study. Farm B provided tiamulin at 1.59 mg/kg of BW as a water treatment for all pigs from d 22 to 42 for treatment of APP.
For this experiment, farm employees were instructed to remove pigs from trial for crippling illness, weight loss, and related compromises in health. On Farms B and C, animals gaining 
0.45 kg during the first 11 and 22 d, respectively, of the study were removed from experimental pens and moved to nonexperimental pens. Despite also being given this same option, Farm A did not remove any slow-growing and unthrifty pigs after initial placement in nursery pens. Farm employees at each nursery recorded the pig’s weight, pen feed consumption, and reason for removal on the date of removal. Medicinal treatment of illness was also recorded, with repetitive or successive treatments of animals noted separately.
Statistical Analyses
Average daily gain, ADFI, and G:F were determined and compared using analysis of variance (GLM) procedures of SAS (SAS Inst., Inc., Cary, NC). These performance measures were calculated using pen weight change, pen feed disappearance, and pig days. The term pig days was determined for each pen by summing the number of days each pig was an occupant of that pen during a given period. Antimicrobial, MOS, farm, sex, and weaning weight block within farm (heavy, medium, or light blocks) were considered main effects. Within each farm, pigs were blocked by BW (light, medium, and heavy on Farms A and C; medium and light on Farm B) and sex and allotted to the four experimental treatments. Weight ranges for the heavy, medium, and light blocks varied among farms. All blocks were not represented within treatment in each nursery (incomplete block design). Pen was the experimental unit, and there were 24, 16, and 24 experimental units providing six, four, and six observations per treatment within Farms A, B, and C, respectively. Treatment differences were compared using the PDIFF option of SAS and considered significant at P < 0.05. All two-way interactions among AM, MOS, farm, and sex, plus the three-way interaction of AM, MOS, and farm, were in the statistical model.
Data for percentage of deaths, removals, and piglets medically treated were compared using the GLIMMIX macro of SAS (Littell et al., 1996) for the analysis of binomial responses using logistic mixed effects models. The models included the fixed effects of treatment, farm, sex, and pen, and the random effect of pens within farms. Least squares means were computed on the logit scale. Estimated percentages (mean, lower 95% confidence limit and upper 95% confidence limit) were calculated by applying the anti-logit link transform of the least squares means and confidence limits. Interactions among the main effects were analyzed. There were no significant (P > 0.50) treatment x farm interactions, so only the overall means for AM and MOS are presented. The incidence of medical treatment was expressed as the number of different animals treated without regard to the number of times each individual animal might have been treated during the course of the experiment.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionImplicationsLiterature Cited |
|---|
Antimicrobial increased (P < 0.001) growth rate during d 0 to 11, d 11 to 42, and the overall period of d 0 to 42 (Table 3). The AM also increased (P = 0.03) feed intake during d 11 to 42 and overall (P < 0.001; Table 4) and improved feed efficiency during d 0 to 11 (P < 0.001; Table 5). These effects occurred on all three farms, but the feed intake response during d 11 to 42 and overall was larger on Farm A than on the other two farms (AM x farm; P = 0.03).
|
|
|
). The MOS improved feed efficiency on Farm C but not on the other two farms (MOS x farm interaction; P = 0.02 for d 11 to 42; P = 0.03 for d 0 to 42; Table 5). During d 11 to 42, MOS increased growth rate more in the absence of AM than in its presence (MOS x AM interaction; P = 0.01). There were no other significant interactions between these two additives.
Relationships of feed additives and indices of health and mortality are shown in Table 6. Percentage of sick pigs treated with injectable medication, percentage of pigs removed from the study for slow growth, and post-weaning survival were not related to MOS or AM addition to nursery diets, or to farm. There were no interactions among main effects of AM, MOS, farm, sex, or pen.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionImplicationsLiterature Cited |
|---|
) and confirm the growth-promoting activity of MOS (Miguel et al., 2003). The 7% increase in growth rate produced by MOS was numerically smaller but not significantly different from the 10% response to AM. The increases in growth rate were associated in some instances with increases in feed intake, and/or with significant improvements in feed efficiency. The growth rate responses to the AM and MOS products tested on the three different farms were not as expected, although the pattern was similar for MOS and AM. The responses were as anticipated on the two commercial farms, in that the animals on Farm A, which had a lower growth rate and relatively high reported disease challenge, had a greater response to either product than animals on Farm B, which had a moderate growth rate and a relatively low reported disease challenge. However, there was a sizeable response to both products on Farm C, the research farm with high historical growth rates. Small responses were expected on Farm C before the start of the experiment because of its excellent herd health status.
Broader surveys of the available information confirm that, in general, responses to both AM (Cromwell, 2001) and MOS (Miguel et al., 2003) are larger when growth rate is slower. The analysis of Miguel et al. (2003), which included the present data, suggested that MOS produces a large response in growth rate of pigs that grow no more than 180 g/d during the first 1 to 2 wk after weaning, but on average, MOS produces little or no response in pigs that grow faster than 180 g/d.
Reasons for our unexpected growth rate response are unclear, but at least two possibilities emerge. First, it might be that the products decreased pathogenic challenges in the intestine, but growth rate was not a good indicator of the degree of that challenge. In other words, myriad environmental and husbandry factors might have supported the rapid growth of pigs on Farm C, despite a significant undocumented pathogenic challenge. Second, the response to either product might have been due to factors other than protection against enteric pathogenic challenges. Gaskins (2001) argued that decreasing the total bacterial population in the intestine would decrease the inflammation of the intestinal tissue, which in turn would decrease the use of energy and amino acids by that tissue and improve productive performance. Either product may have decreased total populations or otherwise altered the microbial ecology to benefit the host. Some commensal bacteria attach to the intestine via Type 1 fimbriae (Weissman et al., 2003). As discussed previously, their populations can be decreased by binding to MOS.
It is possible that the mode of action by which MOS produces a clear increase in growth rate is a direct effect on the immune cells in the gastrointestinal tract via its uptake into M-cells located in the Peyer’s patches on the intestinal surface. Savage and Zakrzewska (1996) reported an increase in plasma IgG and bile IgA in turkeys when fed MOS, and Davis et al. (2004) showed that the addition of MOS to piglet diets can lead to alteration in the populations of leukocytes. Improving the overall intestinal health by binding potential pathogens and by enhancing the animal’s ability to defend against potential antigens by increasing the level of antibody titres, immunoglobulins, and macrophage activity indicates a greater capacity to cope with potential diseases and will ultimately lead to better health and better growth performance.
In contrast to large differences in growth rate among farms, the observed effects of products tested among farms on health status are less clear. Mortality rates were low (<2%) on all farms. The different farms employed very different strategies for managing pigs that became unhealthy. Farm A relied heavily on injectable AM treatments, whereas Farm B removed the unhealthy pigs to a hospital pen and did not provide drug treatments. Farm C used both practices. In all cases, identification of pigs for treatment or removal depended on the judgment of the stockperson on duty, who might have had more or less experience than other employees on that farm or the other farms. Because of these factors, it is not possible to compare the incidence or severity of health problems across farms.
|
Implications |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionImplicationsLiterature Cited |
|---|
|
Footnotes |
|---|
3 Current address: Harvard Business School, Boston, MA 02163.
2 Correspondence: 2209 Anthony Hall (phone: 517-355-8398; fax: 517-432-0190; e-mail: rozeboom{at}msu.edu).
Received for publication December 15, 2004. Accepted for publication July 13, 2005.
|
Literature Cited |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionImplicationsLiterature Cited |
|---|
This article has been cited by other articles:
|
|
 |
|
  W. G. Bergen and G. Wu Intestinal Nitrogen Recycling and Utilization in Health and Disease J. Nutr., May 1, 2009; 139(5): 821 - 825. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. M. C. van der Peet-Schwering, A. J. M. Jansman, H. Smidt, and I. Yoon Effects of yeast culture on performance, gut integrity, and blood cell composition of weanling pigs J Anim Sci, November 1, 2007; 85(11): 3099 - 3109. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
