Salmonella bacteria are ubiquitous in several food production systems and environments.1 Certain Salmonella serotypes are an important cause of foodborne illness in humans and cause over 30% of foodborne deaths associated with bacterial infection in the United States.2 Salmonella bacteria are acquired by the fecal-oral route of transmission, and a few major outbreaks in humans have been associated with consumption of Salmonella-contaminated beef.3–6 Salmonella bacteria can be found on cattle hides and in the feces of cattle,7–9 which may lead to the contamination of beef products during the harvesting process.
Infection with Salmonella bacteria can cause clinical or subclinical illness in cattle that may increase production costs. A paucity of evidence exists for the clinical or subclinical effects of Salmonella infections in feedlot production systems.10–12 Infected animals may not have clinical signs yet may still shed Salmonella bacteria in their feces; therefore, infected cattle without clinical signs may serve as a reservoir for transmission to uninfected cattle or other mammals.10 Preharvest intervention programs that would reduce the prevalence of Salmonella bacteria in the feedlot environment may enhance the safety of beef and, in addition, may improve the health and performance of cattle.13 Although production management factors may affect the prevalence of Salmonella bacteria in the feces of cattle,13,14 it is critical that new strategies are developed to effectively minimize the risks associated with this genus of bacteria.15 A novel vaccine technology that uses SRPs of Salmonella enterica serotype Newport bacteria has been described16 and has the potential to control the prevalence of Salmonella bacteria in several animal species.
This novel vaccine technology makes use of an iron transport mechanism of gram-negative bacteria, which is unique to certain bacterial species but is potentially conserved among Salmonella serotypes.17,18 The method of action of the vaccine is to induce the production of antibodies against SRPs that are located on the outer membrane of certain gram-negative bacteria. Once anti—SRP antibodies bind to the corresponding outer membrane proteins, bacteria will be unable to transport iron across the cell membrane. Because iron is critically important for cell homeostasis, bacteria will die as a result of a lack of iron caused by the inhibition of iron transport mechanisms.19
A commercially available vaccine that has the SRP technology incorporated into its formulation (Salmonella Newport SRP vaccine) is approved for use in cattle for the control of fecal shedding and disease associated with Salmonella enterica subsp enterica serotype Newport. Investigators previously have provided anecdotal reports20,21 that suggest the use of this vaccine is effective for the control of clinical salmonellosis in dairy cattle. Results of 1 studya indicated that the prevalence of Salmonella bacteria in the feces of cull dairy cows that were administered the Salmonella Newport SRP vaccine (7.6%) was significantly lower than that of the prevalence in cull dairy cows that were not administered the vaccine (39.2%). However, investigators in another study21 found no significant effect of the administration of the Salmonella Newport SRP vaccine on subclinical shedding of Salmonella bacteria in the feces of dairy cattle. In yet another study20 in dairy cattle, investigators did not detect a significant effect of the administration of the Salmonella Newport SRP vaccine on fecal shedding, but they did detect a production-enhancing effect in cattle administered this vaccine (milk production was 3% higher and somatic cell counts were 30% lower during the first 30 days of lactation in cattle administered the vaccine than in cattle not administered the vaccine). In addition, SRP technology has been used to reduce the shedding of Escherichia coli O157:H7 in the feces of cattle.22
To the authors' knowledge, studies investigating the effects of the administration of the Salmonella Newport SRP vaccine in beef cattle reared in a feedlot have not been conducted. The purpose of the study reported here was to evaluate the effects of the Salmonella Newport SRP vaccine on cattle health and performance and prevalence of shedding of Salmonella bacteria in the feces of vaccinated feedlot cattle.
Materials and Methods
Cattle—A commercial feedlot (approx capacity, 30,000 cattle) in south central Kansas was selected as the location of the study. A sufficient number of cattle (n = 1,591) to fill 20 pens was purchased for inclusion in the study. Feeder calves that had a mean weight of 227 to 250 kg were procured through typical industry means used by the participating feedlot and arrived in 16 truckloads to the feedlot between October 16 and 25, 2008. Cattle originated from livestock markets and ranches located in Kansas, Oklahoma, South Dakota, and Texas. All cattle were managed in accordance with the feedlot's standard health, feeding, and management protocols that were developed and applied at the discretion of trained feedlot personnel and the consulting veterinarian and nutritionist. Cattle were fed a ration from a series of 4 step-up diets, which primarily consisted of alfalfa hay, distillers grains, and steam-flaked corn, from the time of entry (receiving) to harvest (finishing); the 4 sequential step-up diets had percentages of roughage and concentrate of 46% and 54%, 31% and 69%, 12% and 88%, and 6% and 94%, respectively. Individual animal weights were measured and recorded on days 0 and 21. Pen weights were collected on arrival at the feedlot and at preharvest sampling (within 24 hours of transport to the slaughter facility). This study was approved by the Institutional Animal Care and Use Committee at Kansas State University.
Sample size determination—Sample size estimates were based on the ability to determine a difference in prevalence of fecal shedding of Salmonella bacteria at the time of preharvest sample collection. On the basis of our preliminary data and other reports,12,23,24 we estimated that the mean apparent prevalence of Salmonella bacteria at the time of summer preharvest sample collection would be 40% (range, 0% to 80%) for cattle not vaccinated with the Salmonella Newport SRP vaccine.b We wanted to detect a reduction in apparent prevalence of Salmonella bacteria such that the mean prevalence in pens of cattle vaccinated with the Salmonella Newport SRP vaccine was 25%. Sample size estimates were generated by use of simulation and linear mixed models.c We simulated pen prevalence data as appropriate for the study design, varied the number of pens and samples collected per pen, and analyzed these data by use of mixed models; P values for each simulation analysis were used to generate a graphical output displaying the power to detect hypothesized differences, and the sample size for which the total number of samples and the number of pens for each treatment were displayed. We estimated that 20 pens and 25 samples/pen (at the time of preharvest sampling) would be sufficient to detect a difference, as described, with a type I error rate ≤ 0.05 and a type II error rate < 0.20.
Study design—On arrival at the feedlot, cattle were allocated to pairs (replicates) of study pens. Cattle within each arrival lot were systematically allocated by groups of 3 animals into 2 holding pens until each holding pen contained the appropriate number of cattle to fill the corresponding study pens. Then, pen weights were obtained and cattle were moved to permanent study pens. The allocation process continued until 20 total pens (10 pens/treatment) were filled. Replicates of study pens were adjacently located, and the characteristics (eg, open air and dirt floor) of all pens were typical for the industry standard. For allocation to treatment groups, 1 pen from each replicate was randomly selected by coin toss to be administered the Salmonella Newport SRP vaccineb (vaccinated pen), and the other pen was selected by default to be administered a placebod (control pen).
Cattle were processed (on an individual-pen basis) within 48 hours after arrival at the feedlot. On the initial processing day (day 0), all cattle were administered a dose of a modified-live respiratory virus vaccine,e Mannheimia haemolytica toxoid,f ivermectin,g tilmicosin phosphate,h and 2 mL of the Salmonella Newport SRP vaccine or placebo (SC in the right lateral side of the neck), which was in accordance with the manufacturers' recommendations. On day 21, cattle in the vaccinated pens or control pens were administered a second dose of the Salmonella Newport SRP vaccine or the placebo, respectively; in addition, cattle were administered the first of 2 hormone implantsi and a second modified-live respiratory virus vaccine.j To ensure feedlot processing personnel were unaware of the assignment of cattle to treatment groups, labels of the Salmonella Newport SRP vaccine and placebo vials were covered and coded as vaccine A and vaccine B, respectively. Furthermore, these products had the same fluid color and consistency and were both provided in 100-mL vials. In addition, personnel who administered the treatment were not the same personnel responsible for assessing cattle health and performance during the study; personnel responsible for assessing cattle health and performance also were unaware of treatment group. At approximately 80 days prior to harvest, cattle were administered a third modified-live respiratory virus vaccine,k an external parasiticide,l and a second hormone implant.i
Fecal sample collection—Freshly voided fecal samples (n = 25 fecal samples/pen) were collected from the pen floor on days 0, 60, and 120 and at the preharvest sampling. Each sample was collected manually by use of a clean plastic sleeve, and appropriate precautions were observed to avoid potential contamination of samples by other feces or pen floor material. After each sample was collected, the plastic sleeve was inverted and tied, labeled, and placed in a refrigerated (4°C) cooler until processing at a laboratory.m Laboratory personnel were not aware of the treatment groups during the study period.
Bacterial isolation and serotyping—A previously reported25 standard isolation protocol was used to detect Salmonella bacteria in fecal samples. Ten grams of feces was enriched in 90 mL of tryptic soy brothn in 532-mL stand-up sample bags.o The stand-up sample bags then were incubated at 25°C for 2 hours, 42°C for 6 hours, and 4°C overnight. Samples were agitated and 10 mL of fecal slurry from each sample bag was added to a culture solution that contained 90 mL of tetrathionate brothn and 1.8 mL of iodine.p The culture solution then was incubated at 37°C for 24 hours. After incubation, 1 mL of the culture solution was subjected to immuno-magnetic separation with anti-Salmonella magnetic beads.q The immunomagnetic separation product was adjusted to a final volume of 100 μL with PBS solution/transferred into 10 mL of Rappaport-Vassiliadis broth,m and incubated at 42°C for 16 to 18 hours. The Rappaport-Vassiliadis cultures were vortexed, and 50 μL of each culture was plated onto Hektoen enteric agar platesn and then incubated at 37°C for 24 hours. Three colonies that had morphological characteristics consistent with Salmonella spp were streaked onto blood agars and incubated at 37°C for 24 hours. At least 1 isolate from each sample was tested for the Salmonella polyvalent O antigen as well as serogroups B, C1, C2, D1, D2, and E via slide agglutination.n Isolates, which were presumed to be Salmonella spp on the basis of colony morphology observed on Hektoen enteric agar and agglutination with polyvalent O antisera, were stored at −80°C on cryoprotection beads.t One isolate from each sample was sent to a reference laboratoryu for serotyping.
Statistical analysis—All pen-level cattle health and performance data were collected via the feedlot's operational database. Data were recorded and descriptive analyses were performed by use of a commercially available spreadsheet program.v Exact 95% binomial CIs were calculated for proportions by use of a function included in the spreadsheet programv that outputs the inverse of the cumulative β probability density function for a specified β distribution. All multivariable analyses of fecal shedding and cattle health and performance data were performed by use of a commercial software programw via general and generalized linear mixed models as appropriate for normal and binomial distributions.26 Logistic regression models were used to assess dichotomous outcomes (eg, morbidity, death, and fecal shedding of Salmonella bacteria) among vaccinated and control pens, while including pen within replicate as a random effect. A categorical variable representing sampling times (days 0, 60, and 120 and preharvest) was used when assessing repeated pen measures of fecal prevalence at all 4 sampling periods to allow the investigation of potential time-dependent effects of the vaccine. A first-order autoregressive correlation structure was used, which is a standard approach for repeated measures over equal time periods that allows for power decay of correlations.27 General linear mixed models were used to compare cumulative data for pen-level continuous outcomes (eg, ADG, F:G ratio, and treatment costs) among vaccinated and control pens, while controlling for the lack of independence within a replicate by use of a random intercept model. A value of P < 0.05 was used to indicate significance for all analyses. Fit of a model was assessed by evaluating plots of residuals; for logistic models, the ratio of the deviance of the model to the degrees of freedom of the model was also assessed.
Results
Mean weight of cattle at arrival was 256 kg (n = 1,591). Cattle were allocated to the vaccinated (n = 795) and control (796) groups with 10 pens used to house cattle for each treatment group. Data for health and performance of vaccinated and control cattle were summarized (Table 1). There was no significant (P = 0.80) difference in mean weight at arrival between treatment groups within replicates. The number of cattle per study pen ranged from 61 to 105 (mean, 79.5; median, 68) and 61 to 105 (mean, 79.6; median, 67) for the vaccinated and control pens, respectively. Mean number of days at the feedlot for the vaccinated and control pens was 228.8 (median, 229) and 228.9 (median, 229) days, respectively, and did not differ significantly.
Unadjusted summary data of health and performance outcomes for pens of cattle vaccinated with Salmonella enterica subsp enterica serotype Newport SRP vaccine or a placebo (control pens).*
| Variable | Vaccinated pens (n = 10) | Control pens (n = 10) | P value† |
|---|---|---|---|
| No. of cattle | 795‡ | 796‡ | — |
| Entry weight (kg)§ | 255.6 ± 11.08 | 256.5 ± 11.90 | 0.80 |
| No. of cattle with morbidity∥¶ | 277 (34.8) | 277 (34.8) | 0.99 |
| No. of cattle retreated∥¶ | 17 (2.1) | 15 (1.9) | 0.72 |
| No. of cattle culled for health reasons∥¶ | 9 (1.1) | 4 (0.5) | 0.23 |
| No. of cattle that died∥¶ | 15 (1.9) | 9 (1.1) | 0.22 |
| Case fatality risk (%)# | 5.4 | 3.3 | 0.16 |
| Treatment cost/animal ($)§ | 5.91 ± 3.78 | 5.85 ± 2.46 | 0.93 |
| ADG (kg)§ | 1.38 ± 0.12 | 1.40 ± 0.12 | 0.13 |
| F:G ratio§ | 5.58 ± 0.16 | 5.57 ± 0.18 | 0.74 |
| Cost of gain ($)§ | 1.56 ± 0.03 | 1.58 ± 0.05 | 0.56 |
| Adjusted ADG (kg)§** | 1.41 ± 0.12 | 1.42 ± 0.15 | 0.88 |
| Adjusted F:G ratio§** | 5.46 ± 0.19 | 5.47 ± 0.27 | 0.71 |
| Adjusted cost of gain ($)§** | 1.54 ± 0.04 | 1.54 ± 0.07 | 0.89 |
| Hot carcass weight (kg)§ | 375.6 ± 12.42 | 378.3 ± 14.78 | 0.38 |
| Carcass yield (%)§ | 64.7 ± 0.53 | 64.9 ± 0.39 | 0.12 |
| Carcass price adjustment ($/45.5 kg of carcass weight)§†† | 0.95 ± 0.62 | 1.05 ± 1.07 | 0.70 |
Cattle in the vaccinated and control pens were injected with the Salmonella Newport SRP vaccineb or a placebo,d respectively; vaccine was administered in accordance with the manufacturer's recommendations.
Within a row, there was a lack of significant (P ≥ 0.05) vaccine effects on each outcome; P values were determined by use of multivariable logistic and linear models, which accounted for the paired-pen (replicate) study design.
Cattle were systematically allocated by groups of 3 animals into 2 pens until 20 pens (10 pens/treatment) were filled.
Values are reported as mean ± SD.
Value is reported as the number (%).
Represents a cause for inclusion in an overall culling rate, and these reasons were primarily associated with respiratory tract disease and lameness; furthermore, there were no suspected or confirmed cases of salmonellosis.
Value is based on all causes of morbidity and subsequent death.
Value is adjusted for dead and culled cattle.
Carcass price is an economic index representing carcass premiums and discounts associated with USDA quality grade, yield grade, and several other carcass variables assessed after harvest. Values of premiums and discounts are based on carcass characteristics at the time of harvest.
— = Not determined.
Dates of the 4 fecal sample collection periods were October 17 and 24, 2008 (day 0); December 12, 2008 (day 60); February 13 and 20, 2009 (day 120); and May 28 and June 18, 2009 (preharvest). The single sample collection date for day 60 was to accommodate closure of the participating laboratory for a holiday. The within-pen fecal prevalence of Salmonella bacteria following randomization after arrival (day 0; Figure 1) ranged from 0% (0/25) to 48% (12/25) and differed significantly (P = 0.01) among replicates but did not differ significantly (P = 0.73) between control and vaccinated pens. Overall prevalence of fecal shedding of Salmonella bacteria across all sampling times and treatment groups was 10.6% (211/2,000). Of the 211 Salmonella isolates characterized, most were from serogroups E (n = 166), C1 (20), and C2 (9; Table 2). Predominant serotypes recovered were Anatum (n = 133), Lexington var 15+ (22), Lille (11), Newport (8), and Senftenberg (6).
Summary of Salmonella serotypes isolated from fecal samples of cattle vaccinated with a Salmonella Newport SRP vaccine or a placebo.*
| Serotype† | Serogroup | No. of isolates | ||
|---|---|---|---|---|
| Vaccinated pens | Control pens | Total‡ | ||
| Anatum§ | E | 67 | 66 | 133 (63.0) |
| Lexington var 15+ | E | 11 | 11 | 22 (10.4) |
| Lille | C1 | 3 | 8 | 11 (5.2) |
| Newport | C2 | 4 | 4 | 8 (3.8) |
| Senftenberg | E | 1 | 5 | 6 (2.8) |
| 3,15:z10:– | — | 5 | 0 | 5 (2.4) |
| 6,7:–:1,5 | E | 1 | 3 | 4 (1.9) |
| Tennessee | C1 | 4 | 0 | 4 (1.9) |
| Enteritidis | D1 | 0 | 3 | 3 (1.4) |
| Others and nontypeable | — | 6 | 9 | 15 (7.1) |
| Total | — | 102 | 109 | 211 (100) |
Fecal prevalence of Salmonella bacteria in each vaccinated (n = 10; gray bars) and control (10; black bars) pen within a replicate in samples collected on days 0, 60, and 120 and at preharvest sampling. Cattle in the vaccinated and control pens were injected with 2 mL of the Salmonella Newport SRP vaccineb or a placebo,d respectively; vaccine was administered in accordance with the manufacturer's recommendations. Error bars represent the exact 95% CIs for proportions. Multivariable logistic regression analysis of these data, which was used to account for the paired-pen (replicate) study design, indicated significant differences in prevalence of Salmonella bacteria among sampling times, but no significant difference between treatment groups or a treatment-by-sampling time interaction.
Citation: American Journal of Veterinary Research 72, 2; 10.2460/ajvr.72.2.239
Fecal prevalence of Salmonella bacteria in each vaccinated (n = 10; gray bars) and control (10; black bars) pen within a replicate in samples collected on days 0, 60, and 120 and at preharvest sampling. Cattle in the vaccinated and control pens were injected with 2 mL of the Salmonella Newport SRP vaccineb or a placebo,d respectively; vaccine was administered in accordance with the manufacturer's recommendations. Error bars represent the exact 95% CIs for proportions. Multivariable logistic regression analysis of these data, which was used to account for the paired-pen (replicate) study design, indicated significant differences in prevalence of Salmonella bacteria among sampling times, but no significant difference between treatment groups or a treatment-by-sampling time interaction.
Citation: American Journal of Veterinary Research 72, 2; 10.2460/ajvr.72.2.239
Fecal prevalence of Salmonella bacteria in each vaccinated (n = 10; gray bars) and control (10; black bars) pen within a replicate in samples collected on days 0, 60, and 120 and at preharvest sampling. Cattle in the vaccinated and control pens were injected with 2 mL of the Salmonella Newport SRP vaccineb or a placebo,d respectively; vaccine was administered in accordance with the manufacturer's recommendations. Error bars represent the exact 95% CIs for proportions. Multivariable logistic regression analysis of these data, which was used to account for the paired-pen (replicate) study design, indicated significant differences in prevalence of Salmonella bacteria among sampling times, but no significant difference between treatment groups or a treatment-by-sampling time interaction.
Citation: American Journal of Veterinary Research 72, 2; 10.2460/ajvr.72.2.239
Salmonella bacteria were recovered from all 10 replicates of pens, and cumulative prevalence estimates across all sampling times ranged from 1.5% to 22%. Unadjusted cumulative prevalence of fecal shedding was 10.2% (95% binomial CI, 8.3% to 12.1%) and 10.9% (95% binomial CI, 9.0% to 12.1%) for vaccinated and control pens, respectively. Crude prevalence estimates for each sequential sampling time across all pens were 10.0% (95% binomial CI, 7.5% to 13.0%), 2.4% (95% binomial CI, 1.3% to 4.2%), 29.4% (95% binomial CI, 25.4% to 33.6%), and 0.4% (95% binomial CI, 0.1% to 1.4%). Multivariable analysis indicated significant (P = 0.01) differences in the prevalence of Salmonella bacteria among sampling times. However, there was no significant (P = 0.89) difference between treatment groups and no significant (P = 0.12) treatment-by-sampling time interaction. These effects, or lack thereof, were evident in the display of the raw data for the fecal prevalence of Salmonella bacteria (Figure 1).
Unadjusted summary data of common feedlot cattle health and performance indices were summarized (Table 1). Furthermore, model-adjusted estimates for cumulative incidence risks of adverse health outcomes for all vaccinated and control cattle were calculated (Figure 2). On the basis of multivariable models that accounted for replicates, outcomes did not differ significantly between vaccinated and control pens. Overall morbidity risk in study cattle was 34.8% and ranged from 15.9% to 58.7% within pens; however, there was no significant difference among pens (within replicates) within each treatment group. Illness in the study population was primarily caused by respiratory tract disease and lameness; furthermore, there were no suspected or confirmed cases of salmonellosis. Overall, only 2.1% of the vaccinated cattle and 1.9% of the control cattle were treated for illness > 1 time. No significant difference was detected among pens within each treatment group for the number of cattle requiring treatment > 1 time. Overall mean treatment (ie, medication) costs in vaccinated and control cattle were $5.91 and $5.85/animal, respectively, and no significant difference in mean treatment costs was detected between pens within each treatment group. During the study, 13 cattle were culled because of illness, and no significant difference in culling was detected among pens within each treatment group. Overall mortality risk in the study population was 1.5% and ranged from 0% to 4.9% within pens. Overall mortality risk in pens of vaccinated and control cattle did not differ significantly.
Cumulative incidence risks of adverse health outcomes (morbidity, retreatment, culling, and death) for all vaccinated (n = 10; gray bars) and control (10; black bars) pens of cattle. Risks and corresponding 95% CIs are model-adjusted estimates calculated via logistic regression models that accounted for the paired-pen study design; outcomes did not differ significantly between treatment groups.
Citation: American Journal of Veterinary Research 72, 2; 10.2460/ajvr.72.2.239
Cumulative incidence risks of adverse health outcomes (morbidity, retreatment, culling, and death) for all vaccinated (n = 10; gray bars) and control (10; black bars) pens of cattle. Risks and corresponding 95% CIs are model-adjusted estimates calculated via logistic regression models that accounted for the paired-pen study design; outcomes did not differ significantly between treatment groups.
Citation: American Journal of Veterinary Research 72, 2; 10.2460/ajvr.72.2.239
Cumulative incidence risks of adverse health outcomes (morbidity, retreatment, culling, and death) for all vaccinated (n = 10; gray bars) and control (10; black bars) pens of cattle. Risks and corresponding 95% CIs are model-adjusted estimates calculated via logistic regression models that accounted for the paired-pen study design; outcomes did not differ significantly between treatment groups.
Citation: American Journal of Veterinary Research 72, 2; 10.2460/ajvr.72.2.239
Significant differences for any of the standard measures of feedlot performance were not detected among pens of vaccinated and control cattle (Table 1). On the basis of analysis of pens within replicate, ADG for pens of vaccinated and control cattle did not differ significantly. When adjustments were made for losses related to dead and culled cattle, ADG still did not differ significantly among pens between treatment groups. The F:G ratio, which is calculated as the amount of feed (in kg) delivered divided by the amount of weight gain (in kg), and the adjusted F:G ratio for vaccinated and control cattle did not differ significantly. Overall cost of gain and adjusted cost of gain also did not differ significantly between treatment groups. Further multivariable analysis of pen-level data revealed that mean hot carcass weight, carcass yield, and carcass price adjustment for pens of vaccinated and control cattle did not differ significantly.
Discussion
The study of feeder cattle in a commercial feedlot production system reported here revealed no significant differences between cohorts of vaccinated and control cattle in the prevalence of fecal shedding of Salmonella bacteria or cattle health and performance variables. These findings may have been caused by several factors. First, there may have been a lack of efficacy of the Salmonella Newport SRP vaccine in cattle located in this type of a production setting. Second, there may have been an insufficient number of Salmonella bacteria in this environment, which may have reduced the ability to detect differences between the treatment groups. Last, the use of the Salmonella Newport SRP vaccine in a subset of the population (10 pens) may have reduced the overall exposure of cattle to Salmonella bacteria in the portion of the feedlot where the pens included in this study were located; thus, all cattle in these study pens were not exposed to a sufficient number of Salmonella bacteria to determine vaccine efficacy.
The last explanation, which also may be formally characterized as herd immunity, was suggested by investigators of another study20 as a plausible explanation for the low prevalence of Salmonella bacteria in vaccinated and control cattle in a dairy production system. A significant herd immunity effect on the fecal shedding of E coli O157 in feedlot cattle also has been described.28 In that study,28 unvaccinated feedlot cattle were 59% less likely to have detectable amounts of E coli O157 in their feces when housed with cattle that were vaccinated for E coli O157; although that study28 of E coli O157 was not a pen-level investigation of the shedding of Salmonella bacteria, it suggests that herd immunity may be an important factor when evaluating the effect of a vaccine on fecal bacteria in feedlot production systems. Therefore, it is evident that an evaluation of vaccinated and control cattle located in adjacent pens within a single segment of a feedlot may not be an ideal study design for assessing the efficacy of vaccines against E coli O157 or Salmonella bacteria.
The results of a recent observational studya indicated that the administration of the Salmonella Newport SRP vaccine may reduce the shedding of Salmonella bacteria in the feces of cull dairy cows. However, results of the study reported here revealed similar amounts of shedding of Salmonella bacteria in vaccinated and control cattle and were consistent with the results of 2 randomized controlled trials20,21 conducted in dairy cattle. Investigators in one of these trials20 administered the Salmonella Newport SRP vaccine to 75 cows in a 1,200-cow dairy herd and did not detect significant differences in fecal shedding of Salmonella bacteria between treatment groups; however, they detected significantly higher milk production and lower cumulative somatic cell counts in vaccinated cows. Investigators of the other trial20 administered the Salmonella Newport SRP vaccine to 25% of the mature dairy cows within 2 herds that had a history of salmonellosis and did not detect significant differences in fecal shedding of Salmonella bacteria between vaccinated and control cattle; in addition, measurements of health and performance variables were not made.21 In both trials,20,21 an inability to detect a difference in fecal shedding may have been affected by herd immunity or the relatively small proportion of cattle vaccinated within herds and variability of shedding of Salmonella bacteria in the feces that contributed to a small effective sample size for the potential to detect differences. However, results of these trials20,21 also may indicate a lack of efficacy of the Salmonella Newport SRP vaccine for the reduction of fecal shedding of the diverse Salmonella serotypes found in bovine production systems. In the study reported here, Salmonella Anatum was predominantly detected, whereas other serotypes, which included Salmonella Newport for which the Salmonella Newport SRP vaccine has labeled indications, were detected infrequently or rarely. The distribution and diversity of serotypes that were detected in the present study are similar to the findings of other studies12,15,23,29 on the shedding of Salmonella bacteria in the feces of cattle; however, this diversity in Salmonella bacteria may have affected our ability to detect significant vaccine effects.
Fecal prevalence of Salmonella bacteria in the present study was much lower than we expected for feedlot cattle in this region, particularly at the time of the preharvest sampling (< 1%). In another study12 conducted by our research group and in other studies,23,24,30 a much higher fecal prevalence of Salmonella bacteria in feedlot cattle has been reported; thus, the sample size calculations for the present study were based on an expected prevalence of 40% in nonvaccinated cattle. The lower observed prevalence of Salmonella bacteria in the present study combined with the extreme variability in prevalence among replicates and within replicates over time would have adversely affected our ability to detect significant vaccine effects. However, the prevalence at 120 days (29.4%) in the study reported here was not low, and all but 1 replicate had Salmonella-positive fecal samples, which indicated that cattle were broadly exposed to Salmonella bacteria at some level during the study period. Within-pen prevalence of Salmonella bacteria at the time of that sample collection (day 120) ranged between 0% (0/25) and 80% (20/25), but prevalence estimates among pens within replicates were similar. Given the paired-pen allocation of cattle at arrival, data suggested that shedding of Salmonella bacteria was largely affected by cattle source or factors associated with arrival at the feedlot, even after the cattle had been in the feedlot for several months (Figure 1).
Similar to the objective for many other field studies7,8,12,24 of Salmonella spp, including studies20,21,a in which investigators evaluated the efficacy of the Salmonella Newport SRP vaccine, evaluation of the prevalence of fecal shedding of Salmonella bacteria was a primary objective of the study reported here. Although concentrations of Salmonella organisms within positive fecal samples and prevalence (and concentration) of Salmonella bacteria on the hides of cattle may be important indicators of preharvest food safety,31 we did not measure these indicators in the study reported here. There may have been a significant difference in the concentration of Salmonella bacteria in positive fecal samples even though there were no significant differences in prevalence between vaccinated and control pens. Determination of the presence or concentration of Salmonella bacteria on cattle hides after transport to the slaughter facility also may have revealed differences between cohorts of cattle because evidence suggests prevalence increases during transport.7,32 Surprisingly, only 0.4% (2/500) of samples were positive among all pens immediately prior to harvest; this is arguably the most important potential food safety indicator that was measured. Given those prevalence results, it is extremely unlikely that determining the concentration of Salmonella bacteria within these positive fecal samples (n = 2) would have provided additional useful information for the evaluation of vaccine efficacy This extremely low prevalence, following the much higher prevalence at 120 days, again may be perceived as the potential effect of herd immunity and reduced overall exposure in the study environment or simply may be because of time-dependent effects that were not measured. Preharvest samples were collected in the present study during the summer months, which is presumably the time of year when the shedding of Salmonella bacteria is most common in feedlots.24
Health and performance indicators for cattle included in the present study were typical for this type of cattle and production system. All performance measures were extremely similar among pens in both treatment groups, which suggested that there were no significant vaccine effects. We failed to find evidence that the Salmonella Newport SRP vaccine affected health and performance variables despite the fact that multiple outcome variables were analyzed. To prevent further multiplicity in the analysis, potential carcass effects were assessed by use of a pen-level mean carcass price adjustment, which is an economic index representing carcass premiums and discounts associated with USDA quality grade, yield grade, and several other carcass variables assessed after harvest. If the Salmonella Newport SRP vaccine was considered to be an effective preharvest food safety intervention, it would be important to demonstrate no adverse effects on cattle health and performance. In addition, preharvest interventions that would enhance the safety of beef by reducing the prevalence of Salmonella bacteria in the feedlot environment could theoretically improve cattle health and performance.10
In general, adverse health outcomes were not rare in this study population; thus, potential health effects of the vaccine could have been detected had they existed. However, there were no reported clinical effects consistent with salmonellosis in the cattle of the present study. As described in another study,12 the lack of health and performance effects could have been attributable to an insufficient challenge dose of Salmonella bacteria or to diversity among Salmonella serotypes. Subclinical shedding of Salmonella bacteria has been associated with some adverse health outcomes in feedlot cattle, such as lot-level measures of hospital pen mortality rate and retreatment risks as well as individual-animal case fatality risk for cattle with respiratory tract disease.12 However, the effect on case fatality risk was only detected for cattle shedding serogroup B Salmonella spp and was not associated with overall shedding of Salmonella bacteria. In addition, investigators in that study12 failed to find associations between shedding of Salmonella bacteria and several other common health variables measured in feedlots; the authors suggested that some Salmonella serotypes might be considered commensal bacteria in feedlot cattle.
To our knowledge, the study reported here was the first study conducted to evaluate the effects of the Salmonella Newport SRP vaccine in cattle maintained in a commercial feedlot production setting. Although we did not detect any effects of vaccination with the Salmonella Newport SRP vaccine on the fecal prevalence of Salmonella bacteria or cattle health and performance, we recognize that further investigation of the use of this vaccine in other cattle production settings could provide evidence of vaccine efficacy We also recognize that long-term herd vaccination strategies may need to be considered to affect subclinical fecal shedding in cattle. This may be the reason that a recent observational studya revealed a lower fecal prevalence of Salmonella bacteria in cull dairy cows that had been administered the Salmonella Newport SRP vaccine, compared with that for cull dairy cows that had not been administered the vaccine. However, in 2 other experimental trials20,21 in which investigators allocated dairy cows to receive this same vaccine, no effect of vaccine administration on fecal shedding of Salmonella bacteria was detected. We also recognize that the lower than expected prevalence of Salmonella bacteria in feces combined with the extreme variability in prevalence among replicates and within replicates over time may have adversely affected the ability to detect significant vaccine effects in the study reported here. Furthermore, the present study was conducted in only 1 commercial feedlot, and prevalence and serotypes of Salmonella bacteria vary among feedlots and regions.24 Because the control of Salmonella bacteria in commercial feedlot production systems may enhance food safety and potentially cattle health and performance, further studies are necessary to validate control methods.
ABBREVIATIONS
| ADG | Average daily gain |
| CI | Confidence interval |
| F:G ratio | Feed-to-gain ratio |
| SRP | Siderophore receptor and porin protein |
Loneragan GH, Edrington TS, Nisbet DJ, et al. Salmonella isolated from cattle culled from dairy herds in the Texas High Plains of the USA (abstr), in Proceedings. 12th Symp Int Soc Vet Epidemiol Econ 2009;256.
Salmonella Newport bacterial extract SRP, distributed by AgriLaboratories Ltd, St Joseph, Mo.
SAS, version 9.1, SAS Institute Inc, Cary, NC.
Emulsigen, MVP Laboratories, Omaha, Neb.
Bovishield Gold 5, Pfizer Animal Health, New York, NY.
One Shot, Pfizer Animal Health, New York, NY.
Ivomec, Merial Ltd, Duluth, Ga.
Micotil, Elanco, Greenfield, Ind.
Component TE-S, Vetlife, West Des Moines, Iowa.
Pyramid 2 + Type II BVDV, Fort Dodge Animal Health, Fort Dodge, Iowa.
Titanium 3, AgriLaboratories Ltd, St Joseph, Mo.
Exile, Agripharm, Westlake, Tex.
Preharvest Food Safety Laboratory, College of Veterinary Medicine, Kansas State University, Manhattan, Kan.
Becton Dickinson, Sparks, Md.
Whirl-pak, Nasco, Fort Atkinson, Wis.
1%, Fisher Scientific, Fairlawn, NJ.
Dynal Inc, New Hyde Park, NY.
Sigma-Aldrich, St Louis, Mo.
Remel, Lenexa, Kan.
CryoBeads, Hardy Diagnostics, Santa Maria, Calif.
National Veterinary Services Laboratory, Ames, Iowa.
Microsoft Excel 2007, Microsoft Corp, Redmond, Wash.
Stata, version 10, StataCorp LP, College Station, Tex.
References
- 1.↑
Rodriguez A, Pangloli P, Richards HA, et al. Prevalence of Salmonella in diverse environmental farm samples. J Food Prot 2006; 69: 2576–2580.
- 2.↑
Mead PS, Slutsker L, Dietz V, et al. Food-related illness and death in the United States. Emerg Infect Dis 1999; 5: 607–625.
- 3.
CDC. Outbreak of multidrug-resistant Salmonella Newport—United States, January-April 2002. MMWR Morb Mortal Wkly Rep 2003; 51: 545–548.
- 4.
CDC. Outbreak of Salmonella serotype Typhimurium infection associated with eating raw ground beef—Wisconsin, 1994. MMWR Morb Mortal Wkly Rep 1995; 44: 905–909.
- 5.
CDC. Multistate outbreak of Salmonella Typhimurium infections associated with eating ground beef—United States, 2004. MMWR Morb Mortal Wkly Rep 2006; 55: 180–182.
- 6.
CDC. Outbreak of salmonellosis associated with beef jerky—New Mexico, 1995. MMWR Morb Mortal Wkly Rep 1995; 44: 785–788.
- 7.
Barham AR, Barham BL, Johnson AK, et al. Effects of the transportation of beef cattle from the feedyard to the packing plant on the prevalence levels of Escherichia coli O157 and Salmonella spp. J Food Prot 2002; 65: 280–283.
- 8.
Beach JC, Murano EA, Acuff GR. Prevalence of Salmonella and Campylobacter in beef cattle from transport to slaughter. J Food Prot 2002; 65: 1687–1693.
- 9.
Fegan N, Vanderlinde P, Higgs G, et al. A study of the prevalence and enumeration of Salmonella enterica in cattle and on carcasses during processing. J Food Prot 2005; 68: 1147–1153.
- 10.↑
Smith BP. Salmonellosis in ruminants. In: Large animal internal medicine. St Louis: Mosby Inc, 2002; 775–779.
- 11.
Wray C, Davies RH. Salmonella infections in cattle. New York: CABI Publishing, 2000.
- 12.↑
Alam MJ, Renter DG, Ives S, et al. Potential associations between fecal shedding of Salmonella in feedlot cattle treated for apparent respiratory disease and subsequent adverse health outcomes. Vet Res 2009; 40: 2.
- 13.↑
Smith RA, Griffin DD, Dargatz DA. The risks and prevention of contamination of beef feedlot cattle: the perspective of the United States of America. Rev Sci Tech 1997; 16: 359–368.
- 14.
Losinger WC, Garber LP, Smith MA, et al. Management and nutritional factors associated with the detection of Salmonella sp. from cattle fecal specimens from feedlot operations in the United States. Prev Vet Med 1997; 31: 231–244.
- 15.↑
Callaway TR, Edrington TS, Anderson RC, et al. Gastrointestinal microbial ecology and the safety of our food supply as related to Salmonella. J Anim Sci 2008; 86: E163–E172.
- 16.↑
Stevens ET, Thomson DU. Siderophore receptor and porin protein technology for control of Salmonella and Escherichia coli O157:H7 in cattle, in Proceedings. 38th Annu Conv Am Assoc Bovine Pract 2005;25–29.
- 17.
Payne SM, Finkelstein RA. Critical role of iron in host-bacterial interactions. J Clin Invest 1978; 61: 1428–1440.
- 18.
Neilands JB. Siderophores—structure and function of microbial iron transport compounds. J Biol Chem 1995; 270: 26723–26726.
- 19.↑
Kingsley R, Rabsch W, Stephens P, et al. Iron supplying systems of Salmonella in diagnostics, epidemiology and infection. FEMS Immunol Med Microbiol 1995; 11: 257–264.
- 20.↑
Hermesch DR, Thomson DU, Loneragan GH, et al. Effects of a commercially available vaccine against Salmonella enterica serotype Newport on milk production, somatic cell count, and shedding of Salmonella organisms in female dairy cattle with no clinical signs of salmonellosis. Am J Vet Res 2008; 69: 1229–1234.
- 21.↑
Heider LC, Meiring RW, Hoet AE, et al. Evaluation of vaccination with a commercial subunit vaccine on shedding of Salmonella enterica in subclinically infected dairy cows. J Am Vet Med Assoc 2008; 233: 466–469.
- 22.↑
Thornton AB, Thomson DU, Loneragan GH, et al. Effects of a siderophore receptor and porin proteins-based vaccination on fecal shedding of Escherichia coli O157:H7 in experimentally inoculated cattle. J Food Prot 2009; 72: 866–869.
- 23.
Kunze DJ, Loneragan GH, Platt TM, et al. Salmonella enterica burden in harvest-ready cattle populations from the southern high plains of the United States. Appl Environ Microbiol 2008; 74: 345–351.
- 24.↑
Dargatz DA, Fedorka-Cray PJ, Ladely SR, et al. Prevalence and antimicrobial susceptibility of Salmonella spp isolates from US cattle in feedlots in 1999 and 2000. J Appl Microbiol 2003; 95: 753–761.
- 25.↑
Barkocy-Gallagher GA, Berry ED, Rivera-Betancourt M, et al. Development of methods for the recovery of Escherichia coli O157:H7 and Salmonella from beef carcass sponge samples and bovine fecal and hide samples. J Food Prot 2002; 65: 1527–1534.
- 26.↑
Dohoo I, Martin W, Stryhn H. Chapter 22: mixed models for discrete data. In: Veterinary epidemiologic research. Charlottetown, PE, Canada: AVC Inc, 2003; 499–519.
- 27.↑
Dohoo I, Martin W, Stryhn H. Section 21.5.1: correlation structure. Veterinary epidemiologic research. Charlottetown, PE, Canada: AVC Inc, 2003; 490–493.
- 28.↑
Peterson RE, Klopfenstein TJ, Moxley RA, et al. Efficacy of dose regimen and observation of herd immunity from a vaccine against Escherichia coli O157:H7 for feedlot cattle. J Food Prot 2007; 70: 2561–2567.
- 29.
Cummings KJ, Divers TJ, McDonough PL, et al. Fecal shedding of Salmonella spp among cattle admitted to a veterinary medical teaching hospital. J Am Vet Med Assoc 2009; 234: 1578–1585.
- 30.
Loneragan GH, Brashears MM. Effects of using retention-pond water for dust abatement on performance of feedlot steers and carriage of Escherichia coli O157 and Salmonella spp. J Am Vet Med Assoc 2005; 226: 1378–1383.
- 31.↑
Brichta-Harhay DM, Guerini MN, Arthur TM, et al. Salmonella and Escherichia coli O157:H7 contamination on hides and carcasses of cull cattle presented for slaughter in the United States: an evaluation of prevalence and bacterial loads by immunomagnetic separation and direct plating methods. Appl Environ Microbiol 2008; 74: 6289–6297.
- 32.
Reicks AL, Brashears MM, Adams KD, et al. Impact of transportation of feedlot cattle to the harvest facility on the prevalence of Escherichia coli O157:H7, Salmonella, and total aerobic microorganisms on hides. J Food Prot 2007; 70: 17–21.
