Antimicrobial-resistant bacterial infections from foods of animal origin: understanding and effectively communicating to consumers.

Consumers are increasingly interested in the attributes of the food they consume. This includes what is in the food and how it was raised; and at least some consumers are willing to pay a premium for products with specific attributes. However, the current plethora of labels on the market does not adequately address this issue; rather than providing actionable information, most labels add to the consumer confusion. In addition, there is a tendency toward "absence labels" that can contribute to a negative consumer perception of conventional products that may or may not include the attribute in question. Communication with consumers about the complex and highly technical issue of antimicrobial resistance (AMR) is challenging, and experiences from communication efforts about food safety-related issues demonstrate exactly how challenging this is to communicate clearly. General lessons learned from the science of risk communication can help guide efforts to communicate about the challenging issue of AMR. There are efforts underway to chart out a new approach. A new labeled animal production certification program is under development to provide choice for consumers, while reducing consumer confusion, which mandates antibiotic stewardship practices.

Use of Nonpathogenic Escherichia coli Surrogates as Predictors of the Survival of Nontyphoidal Salmonella, non-O157 Shiga Toxin-Producing Escherichia coli, and Escherichia coli O157 Populations after High Hydrostatic Pressure Processing.

Validated surrogates are a useful tool for studying the response of pathogens to food safety interventions, but better surrogates are needed for studies using high pressure processing. Ground beef (85% lean, 15% fat) was inoculated separately with mixed cultures of Escherichia coli O157, non-O157 Shiga toxin-producing E. coli, nontyphoidal Salmonella, and nonpathogenic E. coli surrogate bacteria. The inoculated ground beef was subjected to high hydrostatic pressures of 200, 400, and 600 MPa for 4, 6, and 8 min at each pressure. High pressure processing at 200 MPa reduced the inoculated populations of the pathogenic bacteria by 0.9 to 1.8 log CFU/g, 400 MPa reduced the inoculated populations by 2.5 to 3.6 log CFU/g, and 600 MPa reduced the inoculated populations by 4.5 to 5.6 log CFU/g. The nonpathogenic E. coli surrogates were more resistant to the effects of high pressure processing than were the inoculated pathogen populations. This finding suggests that the nonpathogenic E. coli surrogates could be used as process control indicators for high pressure processing of ground beef to predict a specific level of pathogen reduction. The surviving populations of the potential surrogate bacteria were proportional to the surviving populations of the pathogenic bacteria. The models allow for an estimation of the potential surviving populations of the pathogenic bacteria based on quantitative results of the populations of the surrogate bacteria.

Validation of a Sequential Hide-On Bob Veal Carcass Antimicrobial Intervention Composed of a Hot Water Wash and Lactic Acid Spray in Combination with Scalding To Control Shiga Toxin-Producing Escherichia coli Surrogates†.

Scalding of hide-on bob veal carcasses with or without standard scalding chemical agents typically used for hogs, followed by an 82.2°C hot water wash and lactic acid spray (applied at ambient temperature) before chilling, was evaluated to determine its effectiveness in reducing Shiga toxin-producing Escherichia coli surrogate populations. A five-strain cocktail of rifampin-resistant, nonpathogenic E. coli surrogates was used to inoculate hides of veal carcasses immediately after exsanguination (target inoculation level of 7.0 log CFU/100 cm2). For carcasses receiving no scalding treatments, spraying with 82.2°C water as a final wash resulted in a 4.5-log CFU/100 cm2 surrogate reduction, and an additional 1.2-log CFU/100 cm2 reduction was achieved by spraying with 4.5% lactic acid before chilling. Scalding hide-on carcasses in 60°C water (no chemicals added) for 4 min in a traditional hog scalding tank resulted in a 2.1-log CFU/100 cm2 reduction in surrogate levels, and a subsequent preevisceration 82.2°C water wash provided an additional 2.9-log CFU/100 cm2 reduction. Spraying a 4.5% solution of lactic acid onto scalded, hide-on carcasses (after the 82.2°C water wash) resulted in a minimal additional reduction of 0.4 log CFU/100 cm2. Incorporation of scalding chemicals into the scald water resulted in a 4.1-log CFU/100 cm2 reduction (1.9 log CFU/100 cm2 greater than scalding without chemicals) in the surrogate population, and the first 82.2°C wash provided an additional 2.5-log CFU/100 cm2 reduction. Application of antimicrobial interventions did not affect the carcass temperature decline during chilling, the pH decline, or the color characteristics of the ribeye or the flank of the bob veal carcasses.

Validation of Thermal Lethality against Salmonella enterica in Poultry Offal during Rendering.

Recent outbreaks of human disease following contact with companion animal foods cross-contaminated with enteric pathogens, such as Salmonella enterica, have resulted in increased concern regarding the microbiological safety of animal foods. Additionally, the U.S. Food and Drug Administration Food Safety Modernization Act and its implementing rules have stipulated the implementation of current good manufacturing practices and food safety preventive controls for livestock and companion animal foods. Animal foods and feeds are sometimes formulated to include thermally rendered animal by-product meals. The objective of this research was to determine the thermal inactivation of S. enterica in poultry offal during rendering at differing temperatures. Raw poultry offal was obtained from a commercial renderer and inoculated with a mixture of Salmonella serovars Senftenberg, Enteritidis, and Gallinarum (an avian pathogen) prior to being subjected to heating at 150, 155, or 160°F (65.5, 68.3, or 71.1°C) for up to 15 min. Following heat application, surviving Salmonella bacteria were enumerated. Mean D-values for the Salmonella cocktail at 150, 155, and 160°F were 0.254 ± 0.045, 0.172 ± 0.012, and 0.086 ± 0.004 min, respectively, indicative of increasing susceptibility to increased application of heat during processing. The mean thermal process constant (z-value) was 21.948 ± 3.87°F. Results indicate that a 7.0-log-cycle inactivation of Salmonella may be obtained from the cumulative lethality encountered during the heating come-up period and subsequent rendering of raw poultry offal at temperatures not less than 150°F. Current poultry rendering procedures are anticipated to be effective for achieving necessary pathogen control when completed under sanitary conditions.

Processes to Preserve Spice and Herb Quality and Sensory Integrity During Pathogen Inactivation.

Selected processing methods, demonstrated to be effective at reducing Salmonella, were assessed to determine if spice and herb quality was affected. Black peppercorn, cumin seed, oregano, and onion powder were irradiated to a target dose of 8 kGy. Two additional processes were examined for whole black peppercorns and cumin seeds: ethylene oxide (EtO) fumigation and vacuum assisted-steam (82.22 °C, 7.5 psia). Treated and untreated spices/herbs were compared (visual, odor) using sensory similarity testing protocols (α = 0.20; ß = 0.05; proportion of discriminators: 20%) to determine if processing altered sensory quality. Analytical assessment of quality (color, water activity, and volatile chemistry) was completed. Irradiation did not alter visual or odor sensory quality of black peppercorn, cumin seed, or oregano but created differences in onion powder, which was lighter (higher L* ) and more red (higher a* ) in color, and resulted in nearly complete loss of measured volatile compounds. EtO processing did not create detectable odor or appearance differences in black peppercorn; however visual and odor sensory quality differences, supported by changes in color (higher b* ; lower L* ) and increased concentrations of most volatiles, were detected for cumin seeds. Steam processing of black peppercorn resulted in perceptible odor differences, supported by increased concentration of monoterpene volatiles and loss of all sesquiterpenes; only visual differences were noted for cumin seed. An important step in process validation is the verification that no effect is detectable from a sensory perspective.

Peri- and Postharvest Factors in the Control of Shiga Toxin-Producing Escherichia coli in Beef.

Certain Shiga toxin-producing Escherichia coli (STEC) strains are important causes of food-borne disease, with hemorrhagic colitis and, in some cases, hemolytic-uremic syndrome as the clinical manifestations of illness. Six serogroups and one serotype of STEC (O26, O45, O103, O111, O121, O145, and O157:H7) are responsible for the vast majority of cases in the United States. Based on recent data for all food commodities combined, 55.3% and 50.0% of the outbreaks of STEC O157 and non-O157 in the United States, respectively, are attributable to beef as a food source. Consequently, the U.S. Department of Agriculture, Food Safety and Inspection Service declared these organisms as adulterants in raw, nonintact beef. In North America, cattle are a major reservoir of STEC strains, with organisms shed in the feces and contaminated hides of the animals being the main vehicle for spread to carcasses at slaughter. A number of peri- and postharvest interventions targeting STEC have been developed, and significant progress has been made in improving the microbiological quality of beef in the past 20 years as a result. However, continued improvements are needed, and accurate assessment of these interventions, especially for non-O157 STEC, would greatly benefit from improvements in detection methods for these organisms.

Evaluation of Escherichia coli biotype 1 as a surrogate for Escherichia coli O157:H7 for cooking, fermentation, freezing, and refrigerated storage in meat processes.

Five Escherichia coli biotype I isolates were compared with E. coli O157:H7 under four common meat processing conditions. The processes that were evaluated were freezing, refrigerating, fermentation, and thermal inactivation. For each study, at least one surrogate organism was not statistically different when compared with E. coli O157:H7. However, the four studies did not consistently show the same isolate as having this agreement. The three studies that involved temperature as a method of controlling or reducing the E. coli population all had at least one possible surrogate in common. In the fermentation study, only one isolate (BAA-1429) showed no statistical difference when compared with E. coli O157:H7. However, the population reductions that were observed indicated the isolates BAA-1427 and BAA-1431 would overestimate the surviving E. coli O157:H7 population in a fermented summer sausage. When all of the data from all of the surrogates were examined, it was found that isolates BAA-1427, BAA-1429, and BAA-1430 would be good surrogates for all four of the processes that were examined in this study. There was no statistical difference noted between these three isolates and E. coli O157:H7 in the refrigeration study. These isolates resulted in smaller population reductions than did E. coli O157:H7 in the frozen, fermentation, and thermal inactivation studies. This would indicate that these isolates would overpredict the E. coli O157:H7 population in these three instances. This overprediction results in an additional margin of safety when using E. coli biotype 1 as a surrogate.

Fluorescent protein-marked Escherichia coli biotype I strains as surrogates for enteric pathogens in validation of beef carcass interventions.

The efficacy of antimicrobial interventions implemented in slaughter establishments to reduce enteric pathogens on beef carcasses should optimally be validated under commercial operation conditions. This study was conducted to identify surrogate organisms for enteric pathogens that could be used to validate beef carcass interventions. The growth, resistance, and attachment properties of nonpathogenic fluorescent protein-marked Escherichia coli strains were compared with those of E. coli O157:H7 and Salmonella strains. Growth curves were obtained based on growth in tryptic soy broth at 37 degrees C. In general, growth parameters were not different among potential surrogates and target pathogens (P > 0.05). Thermal resistance was compared in phosphate-buffered saline (pH 7.4) at 55, 60, and 65 degrees C, and D-values of potential surrogates were not different (P > 0.05) or were higher (P < 0.05) than those of the target pathogens. Acid resistance was tested in phosphate-buffered saline acidified with L-lactic acid at pH 2.5, 3.0, and 3.5, and log reductions (CFU per milliliter) were not different (P > 0.05) among potential surrogates and E. coli O157:H7 strains; however, some Salmonella serotypes were less acid resistant than were surrogates (P < 0.05). The cell surface hydrophobicity was different (P < 0.05) among surrogates and some E. coli O157:H7 strains, but the strength of attachment to beef carcasses was not different (P > 0.05) among all microorganisms. Log reductions (CFU per square centimeter) after application of hot water washes and 2% L-lactic acid sprays on beef carcasses were not different (P > 0.05) among surrogates and pathogens. The nonpathogenic E. coli strains evaluated in this study could be used as surrogates for E. coli O157:H7 and Salmonella to validate hot water and lactic acid interventions on beef carcasses.

Effectiveness of acidic calcium sulfate with propionic and lactic acid and lactates as postprocessing dipping solutions to control Listeria monocytogenes on frankfurters with or without potassium lactate and stored vacuum packaged at 4.5 degrees C.

The safety of ready-to-eat meat products such as frankfurters can be enhanced by treating with approved antimicrobial substances to control the growth of Listeria monocytogenes. We evaluated the effectiveness of acidic calcium sulfate with propionic and lactic acid, potassium lactate, or lactic acid postprocessing dipping solutions to control L. monocytogenes inoculated (ca. 10(8) CFU/ml) onto the surface of frankfurters with or without potassium lactate and stored in vacuum packages at 4.5 degrees C for up to 12 weeks. Two frankfurter formulations were manufactured without (control) or with potassium lactate (KL, 3.3% of a 60% [wt/wt] commercially available syrup). After cooking, chilling, and peeling, each batch was divided into inoculated (four strains of L. monocytogenes mixture) and noninoculated groups. Each group was treated with four different dips: (i) control (saline solution), (ii) acidic calcium sulfate with propionic and lactic acid (ACS, 1:2 water), (iii) KL, or (iv) lactic acid (LA, 3.4% of a 88% [wt/wt] commercially available syrup) for 30 s. Noninoculated frankfurters were periodically analyzed for pH, water activity, residual nitrite, and aerobic plate counts (APCs), and L. monocytogenes counts (modified Oxford medium) were determined on inoculated samples. Surface APC counts remained at or near the lower limit of detection (<2 log CFU per frank) on franks with or without KL and treated with ACS or LA throughout 12 weeks at 4.5 degrees C. L. monoctogenes counts remained at the minimum level of detection on all franks treated with the ACS dip, which indicated a residual bactericidal effect when L. monocytogenes populations were monitored over 12 weeks. L. monocytogenes numbers were also reduced, but not to the same degree in franks made without or with KL and treated with LA. These results revealed the effectiveness of ACS (bactericidal effect) or LA (bacteriostatic effect) as postprocessing dipping solutions to inhibit or control the growth of L. monocytogenes on vacuum-packaged frankfurters stored at 4.5 degrees C for up to 12 weeks.

Prevalence of Salmonella and Campylobacter in beef cattle from transport to slaughter.

The objective of this study was to evaluate the effect of typical production practices during the transport of cattle on the resulting incidence of Salmonella and Campylobacter in the feces, on the hides, and on the carcasses of these cattle and in the environment (trucks, holding pens, and knock boxes). Various factors were evaluated, including the type of animal (feedlot cattle vs. adult pasture cattle), the breed of cattle, the body condition of the animal, the age of the animal, the time of feed and water withdrawal, the contamination level of the transport vehicle at the feedlot or farm, the transport time, the time cattle were held in the holding pen at the plant, and the contamination level of the holding pen. Four groups of each type of animal were sampled on different days. Samples were collected from cattle prior to transport and after transport (rectal and hide swabs) as well as from the carcasses of these cattle. Pre- and posttransit samples were also taken from the transport vehicle and from the holding pen and knock box at the slaughter facility. For feedlot cattle, fecal shedding stayed fairly constant for both organisms before and after transport (3 to 5% for Salmonella and 64 to 68% for Campylobacter). However, the shedding rate for adult cattle increased from 1 to 21% for Salmonella but stayed constant for Campylobacter (6 to 7%). Contamination of hides with Salmonella increased for both animal types from a level of 18 to 20% to a level 50 to 56%. For Campylobacter, the contamination level decreased from 25 to 13% for feedlot cattle but remained unchanged for adult animals (1 to 2%). Nineteen percent of feedlot cattle carcasses and 54% of adult cattle carcasses tested positive for Salmonella, while only2% of feedlot cattle carcasses and none of the adult cattle carcasses tested positive for Campylobacter. Thus, for feedlot cattle, the factors considered in this study did not affect the shedding of either organism but did affect the contamination of hides with both. For adult animals, the factors increased both shedding of and hide contamination with Salmonella only, not Campylobacter.

Serotyping and antibiotic resistance profiling of Salmonella in feedlot and nonfeedlot beef cattle.

As part of a larger study to assess risk factors associated with hide and carcass contamination of beef cattle during transport to slaughter, a total of 281 salmonellae were isolated from 1,050 rectal, hide, carcass, and environmental samples. For feedlot cattle, salmonellae were recovered from 4.0% of rectal samples, 37.5% of hide samples, 19.0% of carcass samples, and 47.4% of environmental samples. For nonfeedlot cattle, salmonellae were recovered from 10.9% of rectal samples, 37.5% of hide samples, 54.2% of carcass samples, and 50.0% of environmental samples. Overall, the five serotypes most commonly associated with feedlot cattle and their environment were Salmonella Anatum (18.3% of the isolates), Salmonella Kentucky (17.5%), Salmonella Montevideo (9.2%), Salmonella Senftenberg (8.3%), and Salmonella Mbandaka (7.5%). The five serotypes most commonly associated with nonfeedlot cattle and their environment were Salmonella Kentucky (35.4%), Salmonella Montevideo (21.7%). Salmonella Cerro (7.5%), Salmonella Anatum (6.8%), and Salmonella Mbandaka (5.0%). Antimicrobial susceptibility testing of all of the isolates associated with feedlot cattle revealed that 21.7% were resistant to tetracycline, compared with 11.2% of the isolates associated with nonfeedlot cattle. None of the other isolates from feedlot cattle were resistant to any of other antimicrobial agents tested, whereas 6.2% of nonfeedlot cattle isolates were resistant to more than four of the antimicrobial agents tested.

Prevalence of Listeria monocytogenes during production and postharvest processing of cabbage.

From November 1999 to May 2000, analyses of 425 cabbage, 205 water, and 225 environmental sponge samples from four cabbage farms with packing sheds and from two packing sheds in the Rio Grande Valley and Uvalde, Tex., were conducted to determine whether Listeria monocytogenes was present. Samples were tested by the Food and Drug Administration method for the isolation of Listeria spp., and confirmed isolates were DNA fingerprinted by repetitive-element sequence-based polymerase chain reaction (rep-PCR). L monocytogenes was isolated from 3% (26 of 855) of the samples. Twenty of these isolates were obtained from cabbage (7 isolates from farms and 13 from packing sheds). Three isolates were from water samples(two from farms and one from a packing shed), and three were from environmental sponge samples of packing shed surfaces. Rep-PCR-generated fingerprints of 21 of the isolates revealed 18 distinctive banding patterns. Four isolates from environmental sponge samples of conveyor belts and from cabbage samples shared identical banding patterns, suggesting common sources of contamination. These identical environmental isolates suggest that contact with packing shed surfaces may be a source of contamination of cabbage. However, the cabbage samples could have arrived contaminated, since they were not washed.

Antimicrobial resistance of Listeria monocytogenes isolated from various cabbage farms and packing sheds in Texas.

Twenty-one isolates of Listeria monocytogenes from cabbage, environmental, and water samples were evaluated for antimicrobial resistance by the disk diffusion method. Ninety-five percent (20 of 21) of the isolates tested were resistant to two or more antimicrobial agents. This finding is significant, since multiresistant strains of Listeria spp. are not commonly found in nature. Eighty-five percent (17 of 20) of the multiresistant strains were resistant to penicillin, and the remaining multiresistant isolates were somewhat sensitive to penicillin. A multiresistant strain showing intermediate sensitivity to penicillin was resistant to gentamicin. One isolate was susceptible to all antimicrobial agents except penicillin. Penicillin- and gentamicin-resistant L. monocytogenes have not previously been reported from human, food, or environmental samples. This study provides evidence of the emergence of multiresistant L. monocytogenes strains, pointing to an increase in the potential threat to human health posed by this pathogen.

Trimming and Washing of Beef Carcasses as a Method of Improving the Microbiological Quality of Meat.

A study to compare procedures and interventions for removing physical and bacterial contamination from beef carcasses was conducted in six carcass conversion operations that were representative of modern, high-volume plants and located in five different states. Treatment procedures included trimming, washing, and the current industry practice of trimming followed by washing. In addition, hot (74 to 87.8°C at the pipe) water washing and rinsing with ozone (0.3 to 2.3 ppm) or hydrogen peroxide (5%) were applied as intervention treatments. Beef carcasses were deliberately contaminated with bovine fecal material at >4.0 log colony-forming units (CFU)/cm2 in order to be better able to observe the decontaminating effects of the treatments. Carcasses were visually scored by 2 to 3 trained personnel for the level of gross contamination before and after treatment. Samples (10 by 15 cm, 0.3 to 0.5 cm thick) for microbiological testing were excised as controls or after application of each procedure or intervention and analyzed for aerobic mesophilic plate counts, Escherichia coli Biotype I counts, and presence or absence of Listeria spp., Salmonella spp., and Escherichia coli O157:H7. Average reductions in aerobic plate counts were 1.85 and 2.00 log CFU/cm2 for the treatments of trimming-washing and hot-water washing, respectively. Hydrogen peroxide and ozone reduced aerobic plate counts by 1.14 and 1.30 log CFU/cm2, respectively. In general, trimming and washing of beef carcasses consistently resulted in low bacterial populations and scores for visible contamination. However, the data also indicated that hot- (74 to 87.8°C at the pipe) water washing was an effective intervention that reduced bacterial and fecal contamination in a consistent manner.