| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
ANIMAL PRODUCTION |
* Alberta Agriculture and Rural Development, 100-5401-1 Ave. S, Lethbridge, Alberta, Canada T1J 4V6; and
Agriculture and Agri-Food Canada, Lethbridge Research Centre, 5403-1 Ave. S, Lethbridge, Alberta, Canada T1J 4B1
 |
Abstract |
|---|
 Top Abstract INTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONLITERATURE CITED |
|---|
Key Words: bacteria monitoring • behavior • feedlot cattle • rope
|
INTRODUCTION |
|---|
|
TopAbstractINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONLITERATURE CITED |
|---|
), Campylobacter spp. (Berry et al., 2006), and Salmonella spp. (Stephens et al., 2007). Because cattle colonized by these bacteria often present no overt clinical symptoms (Berry et al., 2006; Callaway et al., 2006), identification of carriers is difficult. Collection of fecal samples or oral swabs from individual animals immediately before slaughter has proven effective at estimating bacterial levels (Stephens et al., 2007), but cost, handling stress, and carcass bruising (Smith et al., 2004) makes this approach impractical for large-scale food safety monitoring programs. Collection of fecal samples from the pen floor requires that the technician enters the pen, a practice that may disturb the cattle or transmit disease among pens or feedlots. Consequently, alternative means for monitoring the prevalence of zoonotic bacteria in large pens of cattle are required.
Calves have a noted propensity for licking and nibbling (Veissier et al., 1997), and Irwin et al. (2002) introduced the use of manila ropes placed near water troughs or the feed bunk as a means of isolating orally derived bacteria. Orally contacted ropes were subsequently found by Smith et al. (2004) to be a more effective method to detect E. coli O157:H7 in penned cattle than the collection of fecal pats from the pen floor. However, in studies conducted by our laboratory, the utility of ropes for E. coli O157:H7 sampling for detection was variable in feedlot cattle (Stanford et al., 2005a) and of limited use in dairy cattle (Stanford et al., 2005b). The objective of the present study was to examine the extent to which animal factors such as duration in the feedlot and BW along with factors such as climate and number of ropes available influence the expression of these largely oral-based behaviors in feedlot cattle.
|
MATERIALS AND METHODS |
|---|
|
TopAbstractINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONLITERATURE CITED |
|---|
). Study 1—Commercial Feedlot Pens (Commercial Study)
In a commercial feedlot near Picture Butte, Alberta, Canada, 2 observation stations were assembled in a 15-m-wide alleyway used for feed delivery and animal transfer. Observation stations were at a height of 1.6 m and placed so that the ropes monitored were all within 6.5 m of the observation station. A total of 8 pens were monitored containing 78 to 234 cattle per pen (mean 196 ± 19) with a stocking density from 12.1 to 30.0 m2 per animal (mean 15.9). Pens consisted of earthen floors with a concrete apron at the feed bunk and a heated waterer, with straw bedding provided as required during periods of inclement weather. Porosity fences were present (height 3.5 m) as protection from prevailing winds, but pens were otherwise open to the environment.
Pens of cattle that were observed were all finishing feedlot animals with a residence time in the feedlot ranging from 28 to 244 d at the start of the study and BW ranging from 440 to 617 kg. Cattle received feed and water ad libitum, with initial feed delivery occurring 30 min to 1 h before rope placement and a second feed delivery between 1500 and 1600 h. One manila rope (1.2 m) was tied to the railing of each pen adjacent to the feed bunk at a site where ropes in 4 feedlot pens were readily visible from a single observation station.
Pens were monitored for a 9-h period during daylight hours (from 0830 to 1730 h Mountain Daylight Time) over a total of 7 d in late October of 2003, with observations beginning at the time of rope placement. Ropes were removed on a daily basis at the end of the monitoring period. Preliminary data indicated that peak usage occurred shortly after rope placement; consequently, a single observer was assigned to each station for the first 2 h after rope placement. To reduce fatigue, observers worked in shifts of no longer than 4 h. When cattle contacted the rope, observers recorded the ear tag number of the animal, type of contact (oral vs. other), and time of day that the rope was contacted. Weather data, including ambient temperature (maximum and minimum), relative humidity (%), wind speed (m/s), and precipitation (mm), were recorded at the weather station of the Lethbridge Research Centre.
Study 2—Research Feedlot Pens Video Surveillance (Video Study)
Eight feedlot pens (each 270 m2) at the Lethbridge Research Centre, were monitored using digital video cameras (Panasonic WV-CP474, Mississauga, Ontario, Canada) with a varifocal lens (Tamaron 2.8 to 12 mm, Saitama-City, Japan), which was enclosed in environmental housing (Pelco, Clovis, CA). Cameras (1 per pen) were mounted on steel poles at a height of 4.9 m. Video data were collected 1 d/wk for 8 consecutive weeks beginning immediately after entry of the cattle to the feedlot (average age 8 mo, average BW = 252.7 ± 30.6 kg) during December of 2003 and January of 2004 and for 1 d/wk for 6 wk at the end of the finishing period (average age 14 mo, average BW = 541.2 ± 42.8 kg) in June and July of 2004.
Ten Red Angus-cross steers were housed in each pen, and the same animals remained in the pen for the duration of the study. Steers received a forage-based diet for the first 8 wk consisting of 70% barley silage, 25% barley grain, and 5% supplement on a DM basis. Cattle were subsequently transitioned from the forage-based diet to a grain-based diet (85% barley, 10% barley silage, 5% supplement, DM basis) over 21 d and maintained on this diet for the duration of the study. For the first 8 wk of monitoring, the backs of cattle were marked for recognition of individuals with affixed 45-cm white plasticized fabric (Ketchum Manufacturing Inc., Brockville, Ontario, Canada). Subsequently, animals were marked with white nontoxic latex paint. Markings were renewed 24 h before each observation during the monitoring period.
Monitoring began between 0845 and 0930 h, shortly after feed delivery at 0830 h. Manila ropes (1.2 m) were tied to areas of the pen railing adjacent to the feed bunk in areas within the view of the video camera. One or 2 ropes were placed in the pens for a period of 4 h. When 2 ropes were placed in a single pen, they were separated by a distance of 2 m. During the first 4 wk of data collection, 4 pens had 1 rope mounted, whereas 2 ropes were hung in the remaining 4 pens. For the next 4 wk, the number of ropes placed in each pen was crossed over so that 2 ropes were placed in pens that previously had 1, and those pens that previously had 2 were reduced to 1 rope. For the final 6 wk of observation at the end of the finishing period, a similar cross-over was performed between pens with 1 rope and pens with 2 ropes after 3 wk of observation. Weather data [ambient temperature mean, maximum and minimum, relative humidity (%); wind speed (m/s); wind direction; precipitation (mm) as rain or snow] were collected throughout the study at the weather station of the Lethbridge Research Centre.
Video images were captured using Omnicast (Gentec, Dorval, Quebec, Canada) and then analyzed using Observer 5.0 (Noldus, Wageningen, the Netherlands) by 2 technicians. Technicians recorded date, time, animal number, and type of contact with the rope (oral vs. other). Inter- and intraobserver reliability was calculated by both technicians reviewing the data collected in a random pen for 2 h in 2 d as well as having them code behaviors in the same 1-h period on 2 occasions. Cohen’s kappa coefficient (Cohen, 1960) was used to evaluate intra- and interobserver reliability.
Statistical Analyses
Data on type of rope contact (oral vs. other), phase of study (first observation vs. others for commercial study; first 2 observations vs. following 6 observations vs. final 6 observations for video study), day of observation, number of ropes per pen (video study only), and hour of observation were analyzed using the MIXED and GENMOD procedures (SAS Inst. Inc., Cary, NC). Means were compared using the least squares mean linear hypothesis test. Number of ropes, previous rope exposure, day of observation, hour of observation, and interactions were included as fixed terms, with pen as a random effect and day as a repeated measure. The minimum values of compound symmetry were used for selecting the covariance structure for each variable. The impact of weather data on rope usage was determined using the CORR procedure of SAS. For the commercial study, stocking density, days on feed, and average BW were each included in the model and residuals plotted to determine if analyses of covariance were required.
|
RESULTS AND DISCUSSION |
|---|
|
TopAbstractINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONLITERATURE CITED |
|---|
Observations were made during week days due to staff availability for the commercial study. After the first 5 d, abnormally warm, summer-like conditions in late October rapidly changed with the onset of a storm, and temperatures declined by 40°C within the study period (Figure 1). Due to low visibility from blowing snow and a lack of animal activity, observations were suspended for 6 d, before resumption for a 2-d period (Table 1). Additionally, ear tags of animals within 1 of the 8 pens proved largely illegible, and these data were dropped from subsequent analyses. For the video study, data were lost for 2 pens on 2 separate days and for all pens in wk 3 of the finishing period (Table 2). Intra- and interobserver reliability as measured by Cohen’s kappa was 0.77 and 0.84, respectively.
|
|
|
In the commercial study, rope usage remained relatively stable over the first 5 d (Table 1
), whereas in the video study, the number of animals using ropes increased after 2 exposures (Figure 2). Previous exposure to an object may affect behavioral response in livestock as compared with exposure to a novel object (Hemsworth et al., 1996). Observing behavior of pen mates has been shown to reduce fear response in the observing animal (Boissy and Le Neindre, 1990), with cattle also known to mimic the behavior of pen mates (Fell and Clarke, 1993). Consequently, a combination of observing interactions of pen mates with ropes and more rapid interaction after previous exposure (Hagen and Broom, 2003) was possibly responsible for the increased usage of ropes noted in video study cattle after 2 exposures to the ropes. As chewing and licking are natural behaviors for cattle (Veissier et al., 1997; Irwin et al., 2002), only minimal previous exposure to ropes was required to maximize usage, whereas less instinctive behaviors such as running a maze may require up to 20 repetitions before the maze is consistently solved (Hemsworth et al., 1996).
|
; Schwartzkopf-Genswein et al., 2006). The entire feedlot environment would initially be novel to the cattle in the video study; therefore, rope usage may have been reduced during this period due to the large number of novel stimuli that were present within the feedlot environment. Additionally, establishment of a dominance hierarchy among pen mates may have kept animals preoccupied; Raussi et al. (2005) documented that newly mixed pens of heifers had 4 to 11 agonistic interactions per animal per hour. Results of the present study would support the contention that the utility of ropes for monitoring E. coli O157:H7 is impaired in pens of newly arrived cattle at commercial feedlots (Stanford et al., 2005a).
The feedlot environment for cattle is relatively barren, with animal boredom a concern (Pelley et al., 1995). As livestock have been shown to more readily adopt stimulatory behavior in a relatively barren environment (Jones, 1993), the heavy utilization of ropes by cattle in the present studies (as evidenced by an average of more than 100 individual animal contacts per pen per day for the first 5 d of the commercial study, Table 1) can be partly attributed to the lack of stimulation present in the standard feedlot environment. Interaction by cattle with ropes in the video study was equivalent (similar number of interactions per unit time) to that reported by Wilson et al. (2002) in pens of similar size and composition using a moveable scratching/rubbing device. Consequently, supplying ropes for oral contact may be a simple and inexpensive means of environmental enrichment for feedlot cattle, regardless of the efficacy of the ropes for monitoring zoonotic bacteria, although the possibility of ropes transmitting bacteria among feedlot cattle would require evaluation in a future study.
Effect of Age and Size of Cattle on Rope Usage
As cattle progress from weaning and feedlot entry through to the finishing phase, older animals tend to be heavier. It was therefore not possible to separate age- and size-related effects in behaviors associated with rope usage. Yearling cattle are known to be more active and curious than older animals (Murphey et al., 1981), and heavier cattle are generally less active than lighter animals (Veissier et al., 1997). For cattle of the commercial study, increased BW did not alter rope usage because these animals were already in the finishing phase with a minimum BW of 440 kg (Table 1) compared with the 228-kg calves initially observed in the video study (Figure 2). From analysis of residuals in the commercial study, days on feed, stocking density or BW did not affect rope usage, likely due to the relative homogeneity of the commercial cattle observed.
In contrast, during the 9 mo of the video study, calves gained a total of 288 kg, increasing from an initial BW of 253 to a final BW of 541 kg. Accordingly in this study, rope usage markedly declined (P < 0.05) in observations made at the end of the finishing period compared with those made with the same cattle 3 to 6 wk after entry into the feedlot (Figure 2). Potentially, activity levels of these animals could have been influenced by ambient temperature, although the average temperature during the end of the finishing phase in June and July of 2004 was relatively moderate, averaging 19.1°C (data not shown). Temperatures in excess of 25.6°C are required to produce thermal stress in cattle (Lefcourt and Adams, 1998), especially because animals would have been acclimated to warmer temperatures by this time (DeDios and Hahn, 1994).
Although a reduction in rope usage was noted in finishing cattle in the video study relative to the usage of these animals shortly after entering the feedlot, one-third of finishing cattle were sampled in the commercial study in 1 h using 1 rope per pen (Table 3). Other noninvasive monitoring strategies such as collecting a pooled fecal pat from the pen floor may have less sensitivity for detection of zoonotic bacteria than ropes because of difficulties in distinguishing fecal pats from individual animals. Variable distribution of bacteria within fecal pats as well as the rapid dilution of fecal pats with soil and bedding material may also influence the sensitivity of this collection method (Pearce et al., 2004). Consequently, changes in rope usage with age and BW of feedlot cattle, although present, are unlikely to compromise the utility of ropes for monitoring zoonotic bacteria in the crucial period immediately before slaughter.
|
During the snow storm encountered in the commercial study, rope usage was negligible (data not shown), although interactions with the rope partially recovered after 6 d of acclimation to maximum temperatures of 0°C or less (Table 1). Sudden declines in temperature occurring in the fall have been shown to shift circadian rhythms in feedlot cattle (Hahn, 1999), although the behavioral impacts of a rapid 40°C drop in temperature have not previously been reported. A temperature decline of this magnitude would cause considerable stress, promoting huddling and other activities such as seeking shelter (Sato et al., 1991) and reducing the time allotted for nonessential activities such as contact with the ropes. Other sudden climatic events that cause animal stress would also likely impact rope usage and should be considered in bacterial monitoring strategies. In contrast, provided animals are acclimated to ambient temperature, long-term climatic impacts on rope usage appeared to be minimal. Extreme cold (<–25°C) was not found to inhibit rope usage in acclimated feedlot cattle because the cattle were observed to manipulate balls of frozen saliva attached to the ropes during subzero temperatures (K. Stanford, unpublished data).
Although we were able to observe effects of cold stress on rope usage in the present study, heat stress was not experienced due to the combination of ambient temperatures below the upper critical zone, relative humidity <40%, and wind speeds generally >15 km/h (Lefcourt and Adams, 1998). The study of Smith et al. (2004) was conducted in the summer in the Midwestern United States, and although climatic conditions were not noted, it is likely that the strategy of hanging ropes before dusk was used to avoid the inactivity of feedlot cattle noted in times of heat stress (Mitlöhoner et al., 2001). In contrast, licking/grooming and scratching activity of Canadian feedlot cattle in winter and spring months has been shown to occur primarily during daylight (Gonyou and Stricklin, 1984). Consequently, optimal times for placement of ropes will vary and should be synchronized with peak animal activity, such as occurring at feeding, and as influenced by climatic variables.
Number of Ropes and Duration of Rope Placement
In our study, with 1 rope per pen and approximately 60% more animals per pen than in the study of Irwin et al. (2002), mean proportion of animals contacting the rope (physical + oral) was 46% after 2 h of placement (Table 3). Irwin et al. (2002) proposed hanging multiple ropes to improve sampling by cattle and reported a total contact rate (physical + oral) of 55% over a 2-h period with 7 ropes in pens averaging 123 animals. In the present study, sampling of cattle was increased only in the first hour by an average of 1 contact per animal by placing 2 compared with 1 rope in the video study (Figure 3). Although results of the video study, with 10 cattle per pen, would not be completely representative of a large commercial feedlot pen, these results in combination with the rope contact rates noted in the commercial study would not support use of more than 1 rope per pen from the perspective of maximizing the number of animals within a pen that come in contact with the rope. Unless new and inexpensive bacterial detection techniques are developed, the increased expense of collection and analysis would negate any minor sampling advantages that would be obtained from the placement of multiple ropes in terms of defining a pen as positive or negative for specific bacteria. Further work would be required to determine if placement of multiple ropes increases the diversity of zoonotic bacteria collected from a single pen.
|
where cattle exhibited the greatest degree of interest in enrichment devices with the most restricted access.
The proportion of animals sampled per pen in the commercial study increased over the first 6 h of access, but stabilized thereafter (Table 3). Although lengthening the duration the rope was hung may increase the proportion of animals sampled within a pen, there was a significant reduction in the frequency of rope contact after 1 h (Figure 3). Longer intervals between contacts may result in the rope going through cycles of moistening and drying, conditions that may reduce the viability of rope-associated bacteria. Smith et al. (2004) collected ropes in the early morning after allowing night access, a practice that may avoid extreme desiccation of rope and improve isolate recovery. Consequently, appropriate duration of rope access would range from 2 to 6 h, but the effectiveness of recovery of microbial isolates may be influenced by environmental conditions during the sampling period.
Although not part of the present studies, future work is required to assess the impact that environmental conditions have on the recovery of zoonotic bacteria from ropes after oral or physical contact. It is also important to note that physical contact with the rope may not be as effective as prolonged oral contact for isolation of bacteria, and studies to determine the efficiency of isolate transfer by type of contact are also needed.
Animal Reactivity and Other Environmental Stimulants
The cause of the reduced utilization of ropes by cattle in 1 of the pens during the first 7 wk of the video study (Table 2) is unknown but may be due to animal reactivity, increased stimulation, or both in the environment of that pen because it was adjacent to the main handling facility for the entire feedlot. The limited utility of ropes for sampling dairy cattle noted by Stanford et al. (2005b) was thought to be partially related to the environmental complexity of dairy production systems with alternative stimuli reducing the attraction of cattle to ropes. However, because rope usage gradually became equivalent in all pens of the video study, differences in animal reactivity are a more likely explanation. Cattle with little previous human contact are initially more reactive in feedlot situations and spend less time standing still than tamer animals (Fell and Clarke, 1993). Possibly, the 1 pen may have contained a greater proportion of fearful animals that required additional habituation before seeking contact with the rope. Although not directly assessed in the present study, animal excitability is likely to affect voluntary oral and physical sampling using ropes and should be considered in selection of monitoring methodology for zoonotic bacteria.
To maximize the effectiveness of ropes as bacterial sampling devices for groups of feedlot cattle, factors that impact the voluntary utilization of the ropes by the cattle must be known and understood. In situations in which the utility of the ropes would be minimized, other methodologies such as collection of fecal samples from the pen floor could be considered. Results of the present studies would support allowing 2 to 6 h of access to one 1.2-m rope per pen for optimal sampling of feedlot cattle with minimal cost. Placement of 2 ropes per pen only marginally improved sampling of cattle and would increase associated costs. Although rope usage was reduced in heavier finishing cattle as compared with younger calves, 46% of finishing cattle in pens averaging 196 animals made contact with a single rope placed in the pen for a period of 2 h. Sudden climatic changes can dramatically alter rope usage, and calves required 2 wk of adaptation to the feedlot environment before peak rope usage occurred. Animal reactivity may also affect voluntary pathogen sampling because excitable animals may not remain stationary long enough to investigate the ropes. Behavior was found to profoundly affect the utility of ropes for sampling feedlot cattle, and ropes would be most effective in pens where the social environment is stable and cattle are inquisitive, but not highly reactive. Although the current studies identified animal behavioral considerations that will affect the utility of ropes for sampling bacteria in groups of feedlot cattle, the impacts of type of contact (oral vs. physical) and environmental factors such as ambient temperature and relative humidity on survival of bacterial isolates on ropes are unknown and will require future study.
|
Footnotes |
|---|
2 Corresponding author: stanfordk{at}agr.gc.ca
Received for publication April 28, 2008. Accepted for publication August 15, 2008.
|
LITERATURE CITED |
|---|
|
TopAbstractINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONLITERATURE CITED |
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
