Seasonal Presence of Salmonella spp.,Salmonella Typhimurium and Its Monophasic Variant Serotype I 4,[5],12:i:-,in Selected United States Swine Feed Mills

Abstract

This study evaluated the seasonal prevalence and distribution of Salmonella spp., Salmonella enterica serovar Typhimurium (ST) and its monophasic variant 4,[5],12:i:- (STM), in selected swine feed mills across the United States. Eleven facilities were selected for this study and 12 sites were sampled within each mill during fall 2016, early spring 2017, and summer 2017. Samples were evaluated following the USDA-FSIS guidelines for Salmonella isolation and culture positive samples were analyzed by polymerase chain reaction (PCR). A multiplex real-time PCR was used to differentiate ST and STM from other serotypes. Associations between season, mill, and sample site with Salmonella presence were investigated using generalized linear mixed models. Both season (p < 0.007) and mill (p < 0.005) were significantly associated with Salmonella spp. presence. Fall months were associated with a higher Salmonella prevalence (13.2%) compared with early spring and summer. A total of five isolates, among the 383 samples were serotyped as ST and STM. These two serotypes showed a similar seasonal presence throughout the study, being found during fall and summer seasons. These findings demonstrated the seasonal presence of Salmonella spp. in feed mills and the role of these environments as potential pathogen entry route into the human food chain.

Introduction

S almonella is linked to 1.2 million human foodborne illnesses annually. Although the impact on human health alone is enormous, there is also substantial economic repercussion of Salmonella outbreak across the food industry. According to the Centers for Disease Control and Prevention (CDC) (2011), Salmonellosis alone is responsible for $2.3 billion in costs to the human food industry, with nearly 6% of the reported cases associated with pork and/or pork products (Pires et al., 2013; Dickson et al., 2015).

Several studies have demonstrated the role of pigs as Salmonella reservoir (De Knegt et al., 2015). Animals can become infected with foodborne pathogens through contaminated feed (Crump et al., 2002), among other vehicles (Maciorowsky et al., 2006). Feed can be contaminated during its production in the feed mills through contaminated ingredients and/or the environment and equipment (Jones and Richardson, 2004). Animals consume contaminated feed and can then harbor the bacteria without manifesting clinical signs, as asymptomatic carriers, while still shedding the organisms in their feces, promoting a cycle of pathogen spread within the farms and herds (Rostagno and Callaway, 2012).

Recently, Salmonella has been linked to “feedborne outbreaks” (Österberg et al., 2006; Molla et al., 2010). Once the animals are harvested, the processing of the carcass into pork cuts can result in the contact of contaminated gastrointestinal contents from infected pigs to its carcass and the others around it, through fomites such as knives, processing tables, and plant workers (Olsen et al., 2001; Swanenburg et al., 2001; Vieira-Pinto et al., 2005). The resulting contaminated pork products can then be sold to the final consumer and result in human illness if the pork product is not properly cooked to Salmonella inactivation temperatures or if there is cross contamination between the raw pork and other food items or food preparing surfaces (Carrasco et al., 2012).

Although Salmonella spp. contamination in livestock feed is generally low (Li et al., 2012), it is important to understand the locations of entry into the animal feed value chain. For example, raw ingredients can come in contact with foodborne pathogens during transportation and storage (Crump et al., 2002). Once the raw ingredients reach production facilities, microbial contamination can occur while unloading ingredient due to dust creation, pests, and/or during processing and handling of the products (Whyte et al., 2003; Maciorowski et al., 2006). Since Salmonella spp. has been identified as a potential biological hazard in many livestock feeds (Crump et al., 2002; Cochrane et al., 2015), understanding this pathogen's ecological niche and potential preharvest entry routes into the human food chain is critical. Another important parameter to be considered in contamination episodes is pathogen seasonality. This phenomenon is probably the results of different factors, including human behavior, consumption pattern, pathogen prevalence in animal reservoir, and pathogen environmental survival patterns. The likelihood of contamination and pathogen concentration is strongly related to environmental conditions (Liu et al., 2013).

Among the emerging Salmonella serotypes linked to pork product, a monophasic variant of Salmonella Typhimurium (ST), Salmonella enterica 4,[5],12:i:- (STM), has recently caused a large recall from whole roaster hogs (Centers for Disease Control and Prevention [CDC], 2015). Investigations traced the source of contamination to a pork slaughter establishment in Graham, WA, and the potential sources of contamination were identified as the raw pork meat, the inadequate employee handwashing practices, and the poor cleaning conditions of the surfaces and utensils used (CDC, 2015). This serotype is a particular concern because of its known resistance to many common antimicrobials (Moreno Switt et al., 2009). Since 1995, the reported cases of STM have increased in the United States and within recent years STM has been progressively implicated in human disease worldwide (Moreno Switt et al., 2009).

Therefore, this research was undertaken to determine the presence of Salmonella spp., ST and STM, in environmental and feed samples from selected U.S. swine feed mills during fall, spring, and summer months to help elucidate these pathogens ecological niche and potential preharvest entry routes into the human food chain.

Materials and Methods

Sample collection

Eleven feed mills distributed among eight states were selected for this study representative of the main swine production areas within the United States (Magossi et al., 2018). Each of the chosen mills supply feed to swine operations and not to other livestock and poultry species. Six mills produced only mash (nonpelleted meal-based) feed, whereas the other five facilities produced both mash and pelleted feed. Within each feed manufacturing facility, 12 sampling sites were selected, taking into consideration production flow, people traffic, and dust accumulation (Table 1). Samples were collected with a sterile sponge stick presoaked in 10 mL of Buffered Peptone Water (3M, St. Paul, MN) using a 10 × 10 cm sterile template. Surfaces in receiving ingredient pit grating, floors in receiving area, manufacturing area, warehouse, and control/brake room were sampled in triplicates. Single samples were collected from fat intake inlet, exterior of pellet mill, finished product bin boot/product discharge, load-out auger, and broom. Worker shoes samples were collected from both left and right shoe. Finished feed samples were obtained after the pelleting (for pelleted feed) or after mixing (for mash feed) and analyzed following the method described in Chapter 5 of the Bacteriological Analytical Manual (BAM, 2011). All samples were kept under refrigeration conditions and transported to the laboratory. Processing and testing of samples were conducted within 48 h of sampling. Samples were collected for three seasons: fall (October and November 2016), early spring (February and March 2017), and summer (June and July 2017). These sampling points were selected based on Salmonella seasonality graph (CDC, 2015), where an increase in number of cases is usually observed from March to November.

Table 1.

Presence of Salmonella Polymerase Chain Reaction Positive Samples in Feed Mills by Season, Mill ID, Mill Type, and Sampling Site

Variable	n^a (n positives)	Model-adjusted ^b
Variable	n^a (n positives)	Mean prevalence (%)	95% CI	p
Season				0.006
Fall 2016	127 (25)	13.2	5.1–29.7
Spring 2017	130 (9)	3.6	1.1–11.0
Summer 2017	126 (15)	6.7	2.3–17.9
Mill ID				0.005
1-KS	36 (5)	10.8	3.9–26.8
2-CO	36 (1)	1.9	0.2–13.6
3-OK	35 (3)	6.2	1.7–19.9
4-NC	35 (11)	28.4	14.0–49.1
5-IA	34 (6)	13.9	5.4–31.5
6-NC	37 (2)	3.9	0.9–16.1
7-KS	33 (13)	37.4	19.9–58.8
8-MN	33 (3)	6.7	1.8–21.4
9-IA	33 (0)	0.0	0.0–100
10-IN	33 (5)	11.9	4.1–29.6
11-IL	38 (0)	0.0	0.0–0.0
Mill type				0.952
Mash	204 (27)	6.8	1.7–23.5
Pelleted	179 (22)	7.3	1.8–25.6
Sample site				0.170
1. Receiving ingredients pit gratin	33 (7)	14.7	4.0–41.5
2. Fat intake inlet	32 (1)	1.6	0.2–14.7
3. Pellet mill	16 (2)	7.5	1.1–36.1
4. Discharge bin boot	33 (3)	5.0	1.0–21.9
5. Load-out auger	32 (0)	0.0	0.0–100
6. Finished feed	39 (2)	3.0	0.5–16.6
7. Control room floor	33 (6)	11.9	3.1–36.7
8. Receiving area floor	33 (9)	20.9	6.3–50.7
9. Manufacturing area floor	33 (7)	14.7	4.0–41.5
10. Warehouse area floor	33 (3)	4.7	1.0–21.9
11. Worker shoes	33 (6)	11.9	3.1–36.7
12. Broom	33 (3)	5.0	1.0–21.9
Total	383 (49)

Number of samples collected and analyzed (number of PCR+ Salmonella) per variable considered in this study.

Model-adjusted prevalence estimates from univariable models evaluating the association between each variable with the presence of Salmonella spp.

PCR+, polymerase chain reaction positive.

Culture- and molecular-based analysis

Samples were analyzed following the USDA-FSIS laboratory guidelines for the isolation and identification of Salmonella from meat, poultry, pasteurized egg and catfish products, and carcass and environmental sponges (USDA Food Safety and Inspection Service, 2014). For feed samples 50 g were used for analysis following the Bacterial Analytical Methods, Chapter 5 (BAM, 2011). Sponges were pre-enriched with 60 mL of Buffered Peptone Water (BD Difco, Sparks, MD) at 35°C ± 2°C for 24 ± 2 h, followed by an enrichment in both tetrathionate broth (BD Difco) and Rappaport-Vassiliadis broth (BD Difco) at 42°C for 24 h, and then plated for isolation on Brilliant Green Agar (BD Difco) and Xylose Lysine Tergitol 4 Agar (BD Difco). Phenotypically Salmonella positive samples were then submitted to biochemical tests in Lysine Iron Agar test (BD Difco) and Triple Sugar Iron Agar (BD Difco). Samples that tested positive based on culture were subjected to a real-time polymerase chain reaction (PCR) assay adapted from Bai et al. (2018) targeting the invA gene present in S. enterica and Salmonella bongori. Total reaction volume was 25 μL (12.5 2 × IQ Multiplex Power mix; Bio-Rad, Hercules, CA): 1 μL of each primer (Bioresearch Technologies, Petaluma, CA), 0.5 μL of probe (Bioresearch Technologies), 10 μL of nuclease-free molecular biology grade water (Integrated DNA Technologies, Coralville, IA), and a pick from a colony as the DNA template. Running conditions: an initial denaturation step of 95°C for 10 min, followed by 40 cycles of 95°C for 15 s and 60°C for 50 s. Samples having a Ct value <40 were considered PCR positive (PCR+).

Multiplex PCR

PCR+ isolates were further analyzed by a multiplex PCR assay to differentiate ST and STM from other serotypes. The protocol described by Prendergast et al. (2013) was followed with minor modifications. A pick from a colony was transferred directly from an agar plate, with a pipette tip, to the PCR mixture without any treatment. The reaction was carried out in a final total volume of 25 μL, containing 1 μL of primer mix (0.4 μM of each primer) (Bioresearch Technologies), 0.5 μL (0.2 μM) of each probe (Bioresearch Technologies), 12.5 μL of 2 × IQ Multiplex Power mix (Bio-Rad), and 10 μL of nuclease-free molecular biology grade water (Integrated DNA Technologies). Three sets of primers and probes were used in the assay and the targeted genes were fliC (present in ST and STM), fljB 1,2 (present in ST), and fliB/IS200 (present in ST and STM) (Prendergast et al., 2013). The PCR was carried out in a CFX96 thermocycler (Bio-Rad), with initial denaturation step of 94°C for 2 min, followed by 40 cycles of 95°C for 20 s and 60°C for 90 s. Reaction was considered positive when Ct values ≤40. Samples were characterized as ST if expressing all three genes (fliC, fljB 1,2, and fliB/IS200) and STM if expressing both fliC and fliB/IS200 genes.

Statistical analysis

Generalized linear mixed models were fitted in SAS 9.4 (SAS Institute, Inc., Cary, NC) using the GLIMMIX procedure. Binary distribution, logit link, Laplace approximation, and ridge-stabilized Newton–Raphson algorithm were used. The outcome consisted of the presence of Salmonella spp. in environmental samples as determined by the PCR test (dichotomous: positive vs. negative). Independent variables included season (fall, spring, and summer), mill ID (each individual mill received an ID consisting of a number from 1 to 11 and a state abbreviation), mill type (divided into mills producing mash or pelleted feed), and sample site (numbered from 1 to 12, representing the sites on Table 1). When at least one of the subsamples of floor and worker shoes tested positive (PCR+), sample sites were considered positive. Depending on the fixed effect evaluated in the univariable model, we incorporated a random intercept for feed mill nested within state (except when evaluating feed mill or state as fixed effect), to account for the hierarchical structure of the study. Random effects considered for the univariable models were season, state, mill ID, mill type, month, and season. Mean probabilities and their 95% confidence intervals were computed (Table 1).

Results

A total of 383 environmental and feed samples were collected from 11 feed mills during the three seasons. From the total isolates, 49 (12.8%) were Salmonella PCR+: 2 isolates (5.1%) were identified in feed, whereas the other 47 (13.7%) in equipment and/or on surfaces. Based on the univariable models, season (p < 0.007) and mill ID (p < 0.005) were significantly associated with the presence of Salmonella spp., whereas mill type (p > 0.952) and sample site (p > 0.170) were not (Table 1). Samples collected during fall months had a significantly higher mean prevalence (13.2%) of Salmonella compared with samples collected during early spring (3.6%) or summer (6.7%) (Table 1). Nine of the 11 feed mills had at least 1 Salmonella spp. Within-mill prevalence across the three visits ranged from 1.9% to 37.5%. Facility 4 and 7 had the highest mean prevalence with 28.5% and 37.5%, respectively, whereas mills 9 and 11 had no test positive samples (Table 1). As shown in Table 1, a higher mean prevalence of Salmonella spp. was observed in sites corresponding to receiving area floor (20.9%), manufacturing area floor and receiving ingredients pit gratin (14.7%), followed by control room floor and worker shoes (11.9%). Interaction between significant fixed effects (mill ID* season) was tested (p = 0.999) and the random effects month and state were considered confounded with season and mill ID, respectively. When an effect was considered fixed, it was removed from the list of random effects and the factors were analyzed independently. Therefore, mill ID (p < 0.003) and season (p < 0.005) were significantly associated with prevalence of Salmonella in our multivariable model.

A multiplex PCR was designed to identify ST and STM among the 49 PCR+ Salmonella isolates. A total of two ST and three STM were identified by the multiplex PCR. Both ST isolates originated from mill 5 and were recovered from the receiving area floor during summer. One STM isolate came from mill 1 and was identified in the control room floor during summer. The other two STM isolates were found in mill 10 during fall from the receiving ingredients pit gratin and receiving area floor. These results show that the ST and STM isolates are more prevalent during fall and summer. In addition, the sample sites where ST and STM were recovered matched the highest percentage of PCR+ samples.

Discussion

Governmental agencies have recently implemented program for the surveillance of Salmonella in animal feed (Li et al., 2012). Human salmonellosis has been proven to exhibit seasonal variation: higher prevalence in warmer months and lower in colder months (D'Souza et al., 2004; Pangloli et al., 2008; Ravel et al., 2010). Also in our study, we observed a higher PCR+ samples prevalence in fall and summer seasons (Table 1), which is consistent with the data reported from other studies (Ravel et al., 2010; Jahne et al., 2015). During warmer months' people tend to walk around the facility more often, go outside, and keep doors and windows open for air circulation. This behavior leaves the mill more susceptible to the entrance and spread of microorganisms. Conversely, during colder months people have the tendency of remaining inside and keep doors and windows closed. Other factor contributing to the seasonality of bacterial contamination is represented by airborne transmission of Salmonella from the use of swine manure as fertilizer during the fall (Jahne et al., 2015). Another important factor to be considered for Salmonella surveillance and intervention design is that pig does not shed constantly: irregular shedding was observed as a function of time, cohorts, and pigs (Baptista et al., 2010). Until now U.S. approach to decrease and control the risk of Salmonella has been primarily focused during slaughter and processing. Recent studies have highlighted the importance of management practices, environment, biosecurity plans at preharvest level (Funk et al., 2001; Pires et al., 2013). Interventions on farm are becoming more stringent with the goal of reducing the preharvest prevalence of Salmonella-positive swine.

In our study we observed a significant association between the feed mill ID and the prevalence of Salmonella. Differences in management, geographical location, hygiene practices, quality of incoming raw ingredients, volume of feed produced, number of workers, and time the facility has been operational are all important variables for pathogen presence as described by Cochrane et al. (2015). Birds and bird feces were found in some facilities, highlighting the vulnerability of these production environments to pests, wildlife, weather condition, and human/vehicle traffic (Whyte et al., 2003; Torres et al., 2011). Two different types of mills were included in our study: one producing mash and the other pelleted feed. Facilities were structurally different: extra equipment was present for the pelleting process (conditioner, extruder, pellet mill, and cooler). No significant differences were observed between mill types, probably due to same amount of dust accumulation and human flow as vehicles of microbial spread around the facility. Production flow and plant design might also play a role in preventing microbial introduction and recontamination of finished feeds (Whyte et al., 2003). Research studies showed that raw grain ingredients and transporting trucks were the vehicle of contamination into the mill facilities (Fedorka-Cray and Hogg, 1997; Binter et al., 2011). As in our study, a high number of PCR+ samples were found in the receiving ingredient pit gratin and receiving area floor. Because Salmonella can survive for long periods of time in dry and hostile environments, in our analysis we considered worker shoes as a potential microorganism reservoir (Table 1). Amass et al. (2000) and Otake et al. (2002) proved that shoes can carry biological hazards, such as porcine reproductive virus and respiratory syndrome virus. Therefore, the worker shoes can represent a vehicle for pathogen spread around the mill. Not surprisingly, in our study, the control room and manufacturing floor area with the highest human flow showed a high percentage of PCR+ samples.

Our main goal was to investigate the presence of Salmonella in feed mill during processing and have an understanding of pathogen presence in these environments.

The risk of salmonellosis from feed is difficult to quantify due to inconsistent data, sampling constraints, and lack of epidemiological information (Crump et al., 2002). Limited practices have been implemented, even if these facilities have been recognized as potential source of infections in different occasions (Podolak et al., 2010; Rostagno and Callaway, 2012). Nevertheless, in our study we reported that two feed samples resulted in PCR+ (5.1%) and they came from mills 4 and 7. Both facilities had the highest mean prevalence of Salmonella, 28.4% and 37.4%, respectively. Based on these observations we can hypothesize that bacterial prevalence in feed mill environments can be connected with a risk of cross contamination of finished feed. These results are consistent with the FDA surveillance findings from 2007 to 2009, where 5.6% of finished feed was contaminated with Salmonella (Li et al., 2012). Even if a decrease in prevalence has been observed during the past decade, pathogen presence in feed can still represent one of the potential sources of salmonellosis in human (Li et al., 2012).

The final goal of our study was to identify, among PCR+ samples, the presence of ST and STM. These two serotypes were found most commonly during fall and summer. STM, along with ST, is one of the most commonly found serotype in humans, swine, and pork products in recent decades (Moreno Switt et al., 2009; Hauser et al., 2010). STM showed resistance to many antibiotic drugs generally used to treat human patients.

The data gathered in this study demonstrate the presence of Salmonella spp., ST and STM, in U.S. feed mills. These observations can contribute to the implementation of biosecurity plans and other preventative strategies in these processing facilities. For a complete assessment of Salmonella prevalence in the U.S. mills, further studies will be necessary to extend the geographical range of targeted states and increase the number of mills participating in the study (more mills per state).

Conclusions

Our study demonstrated the presence of Salmonella in feed mills across the United States. A seasonal pattern was observed with higher pathogen prevalence in fall and summer. A total of five ST and STM isolates were found among PCR+ samples. Hygiene, management, production flow, and cross contamination within facility are all important factors previously linked with pathogen contamination in mills. We found that both the mill and the season were significantly associated with Salmonella prevalence. Only one finished feed sample was collected per visit, which could be a source of underestimation of the true contamination status of the production. Nevertheless, these findings contribute to understand Salmonella ecological niche in the animal feed processing environment. Antibiotic resistance pattern, genetic relatedness, and origin of those isolates should also be investigated to evaluate the risk of preharvest microbial entry routes into the human food chain. Moreover, it is important to emphasize that this study reflects the population of selected mills across the United States.

Footnotes

Acknowledgments

The authors thank the National Pork Board for the financial support of this project (NPB no. 16-192), Professor Martin Wiedmann from the Department of Food Science at Cornell University, and Dr. Lisa Gorski from the Produce Safety and Microbiology Research Unit USDA in Albany (CA) for providing the Salmonella Typhimurium monophasic variant 4,[5],12:i:- isolates.

Disclosure Statement

No competing financial interests exist.

References

Amass

, Stevenson

, Anderson

, Grote

, Dowell

, Vyverberg

, Kanitz

, Ragland

. Investigation of people as mechanical vectors for porcine reproductive and respiratory syndrome virus. J Swine Health Prod, 2000; 8:161–166.

Bacterial Analytical Manual. Chapter 5: Salmonella. 2011. Available at: https://www.fda.gov/Food/FoodScienceResearch/LaboratoryMethods/ucm070149.htm, accessed April 3, 2016 .

Bai

, Trinetta

, Shi

, Noll

, Magossi

, Zheng

, Porter

, Cernicchiaro

, Renter

, Nagaraja

. A multiplex real-time PCR assay, based on inv A and pag C genes, for the detection and quantification of Salmonella enterica from cattle lymph nodes. J Microbiol Methods, 2018; 148:110–116.

Baptista

, Dahl

, Nielsen

. Factors influencing Salmonella carcass prevalence in Danish pig abattoirs. Prev Vet Med, 2010; 95:231–238.

Binter

, Straver

, Häggblom

, Bruggeman

, Lindqvist

P-A

, Zentek

, Andersson

. Transmission and control of Salmonella in the pig feed chain: A conceptual model. Int J Food Microbiol, 2011; 145:S7–S17.

Carrasco

, Morales-Rueda

, García-Gimeno

. Cross-contamination and recontamination by Salmonella in foods: A review. Food Res Int, 2012; 45:545–556.

Centers for Disease Control and Prevention (CDC). Multistate Outbreak of Multidrug-Resistant Salmonella I 4,[5],12:i:- and Salmonella Infantis Infections Linked to Pork (Final Update). 2015. Available at: https://www.cdc.gov/salmonella/pork-08-15/, accessed April 20, 2018 .

Centers for Disease Control and Prevention (CDC). National Enteric Disease Surveillance: Salmonella Surveillance Overview. 2011. Available at: https://www.cdc.gov/nationalsurveillance/salmonella-surveillance.html, accessed March 8, 2018 .

Cochrane

, Dritz

, Woodworth

, Stark

, Huss

, Cano

, Thompson

, Fahrenholz

, Jones

. Feed mill biosecurity plans: A systematic approach to prevent biological pathogens in swine feed. J Swine Health Prod, 2015; 24:154–164.

10.

Crump

, Griffin

, Angulo

. Bacterial contamination of animal feed and its relationship to human foodborne illness. Clin Infect Dis, 2002; 35:859–865.

11.

De Knegt

, Pires

, Hald

. Attributing foodborne salmonellosis in humans to animal reservoirs in the European Union using a multi-country stochastic model. Epidemiol Infect, 2015; 143:1175–1186.

12.

Dickson

, Hurd

, Rostagno

. Salmonella in the pork production chain. National Pork Board/American Meat Science Association Fact Sheet. 2015. http://articles.extension.org/pages/27413/salmonella-in-the-pork-production-chain, accessed December 21, 2018 .

13.

D'Souza

, Becker

, Hall

, Moodie

KBA

. Does ambient temperature affect foodborne disease?. Epidemiology, 2004; 15:86–92.

14.

Fedorka-Cray

, Hogg

, Gray

, Lorenzen

, Velasquez

, Von Behren

. Feed and feed trucks as sources of Salmonella contamination in swine. J Swine Health Prod, 1997; 5:189–193.

15.

Funk

, Davies

, Nichols

. Longitudinal study of Salmonella Enterica in two, three-site production systems. Vet Microbiol, 2001; 83; 45–60.

16.

Hauser

, Tietze

, Helmuth

, Junker

, Blank

, Prager

, Rabsch

, Appel

, Fruth

, Malorny

. Pork contaminated with Salmonella enterica serovar 4,[5],12:i:-, an emerging health risk for humans. Appl Environ Microbiol, 2010; 76:4601–4610.

17.

Jahne

, Rogers

, Holsen

, Grimberg

. Quantitative microbial risk assessment of bioaerosols from a manure application site. Aerobiologia, 2015; 31:73–87.

18.

Jones

, Richardson

. Salmonella in commercially manufactured feeds. Poult Sci, 2004; 83:384–391.

19.

, Bethune

, Jia

, Lovell

, Proescholdt

, Benz

, Schell

, Kaplan

, McChesney

. Surveillance of Salmonella prevalence in animal feeds and characterization of the Salmonella isolates by serotyping and antimicrobial susceptibility. Foodborne Pathog Dis, 2012; 9:692–698.

20.

Liu

, Hofstra

, Franz

. Impacts of climate change on the microbial safety of pre-harvest leafy green vegetables as indicated by Escherichia coli O157 and Salmonella Spp. Int J Food Microbiol, 2013; 163:119–128.

21.

Maciorowski

, Herrera

, Kundinger

, Ricke

. Animal feed production and contamination by foodborne Salmonella . J Verbrauch Lebensm, 2006; 1:197–209.

22.

Magossi

, Cernicchiaro

, Dritz

, Houser

, Woodworth

, Jones

, Trinetta

. Evaluation of Salmonella presence in selected United States feed mills. Microbiologyopen, 2018; e00711.

23.

Molla

, Sterman

, Mathews

, Artuso-Ponte

, Abley

, Farmer

, Rajala-Schultz

, Morrow

, Gebreyes

. Salmonella enterica in commercial swine feed and subsequent isolation of phenotypically and genotypically related strains from fecal samples. Appl Environ Microbiol, 2010; 76:7188–7193.

24.

Moreno Switt

, Soyer

, Warnick

, Wiedmann

. Emergence, distribution, and molecular and phenotypic characteristics of Salmonella enterica serotype 4,5,12:i:-. Foodborne Pathog Dis, 2009; 6:407–415.

25.

Olsen

A-M

, Jensen

, Dahl

, Christensen

Reduction in level of Salmonella on swine carcasses after slaughter without splitting the head. In: Proceedings of the 4th International Symposium on the Epidemiology and Control of Salmonella and Other Food Borne Pathogens in Pork, Leipzig, 2001:124–126.

26.

Österberg

, Vågsholm

, Boqvist

, Lewerin

. Feed-borne outbreak of Salmonella Cubana in Swedish pig farms: Risk factors and factors affecting the restriction period in infected farms. Acta Vet Scand, 2006; 47:13–22.

27.

Otake

, Dee

, Rossow

, Deen

, Joo

, Molitor

, Pijoan

. Transmission of porcine reproductive and respiratory syndrome virus by fomites (boots and coveralls). J Swine Health Prod, 2002; 10:59–65.

28.

Pangloli

, Dje

, Ahmed

, Doane

, Oliver

, Draughon

. Seasonal incidence and molecular characterization of Salmonella from dairy cows, calves, and farm environment. Foodborne Pathog Dis, 2008; 5:87–96.

29.

Pires

, Funk

, Bolin

. Longitudinal study of Salmonella shedding in naturally infected finishing pigs. Epidemiol Infect, 2013; 141:1928–1936.

30.

Podolak

, Enache

, Stone

, Black

, Elliott

. Sources and risk factors for contamination, survival, persistence, and heat resistance of Salmonella in low-moisture foods. J Food Prot, 2010; 73:919–1936.

31.

Prendergast

, Hand

, Nί Ghallchóir

, McCabe

, Fanning

, Griffin

, Egan

, Gutierrez

. A Multiplex real-time PCR assay for the identification and differentiation of Salmonella enterica serovar Typhimurium and Monophasic serovar 4,[5],12:i:–. Int J Food Microbiol, 2013; 166:48–53.

32.

Ravel

, Smolina

, Sargeant

, Cook

, Marshall

, Fleury

, Pollari

. Seasonality in human salmonellosis: Assessment of human activities and chicken contamination as driving factors. Foodborne Pathog Dis, 2010; 7:785–794.

33.

Rostagno

, Callaway

. Pre-harvest risk factors for Salmonella enterica in pork production. Food Res Int, 2012; 45:634–640.

34.

Swanenburg

, Urlings

HAP

, Snijders

JMA

, Keuzenkamp

, Vana Knapen

. Salmonella in slaughter pigs: Prevalence, serotypes and critical control points during slaughter in two slaughterhouses. Int J Food Microbiol, 2001; 70:243–254.

35.

Torres

, Piquer

, Algarra

, de Frutos

, Sobrino

. The prevalence of Salmonella enterica in Spanish feed mills and potential feed-related risk factors for contamination. Prev Vet Med, 2011; 98:81–87.

36.

USDA Food Safety and Inspection Service. Isolation and Identification of Salmonella from Meat, Poultry, Pasteurized Egg, and Catfish Products and Carcass and Environmental Sponges. 2014. Available at: https://www.fsis.usda.gov/wps/wcm/connect/700c05fe-06a2-492a-a6e1-3357f7701f52/MLG-4.pdf?MOD=AJPERES, accessed September 13, 2016 .

37.

Vieira‐Pinto

, Temudo

, Martins

. Occurrence of Salmonella in the ileum, ileocolic lymph nodes, tonsils, mandibular lymph nodes and carcasses of pigs slaughtered for consumption. J Vet Med, 2005; 52:476–481.

38.

Whyte

, McGill

, Collins

. A survey of the prevalence of Salmonella and other enteric pathogens in a commercial poultry feed mill. J Food Saf, 2003; 23:13–24.