Combining enrichment with multiplex real-time PCR leads to faster detection and identification of Campylobacter spp. in food compared to ISO 10272-1:2017.

Conventional protocols for the detection of Campylobacter from foods are laborious and time-consuming. This research describes an alternative procedure (EMRT-PCR) for the detection of Campylobacter from food by combining ISO 10272-1:2017 enrichment in Bolton broth (BB) with a multiplex real-time (MRT-) PCR assay. Species differentiation was done by targeting C. jejuni (mapA), C. coli (ceuE), and both species (cje). The detection limit of the MRT-PCR assay was 4.5 and 5.5 log10 cfu/ml in BB and BB containing chicken skin, respectively. A Monte Carlo simulation was conducted to predict the probability that campylobacters reach the MRT-PCR detection threshold throughout enrichment in BB, and results suggested that cold-stressed campylobacters could reach the detection limit after 40 h of enrichment (p = 0.99). As a proof of principle, 23 naturally contaminated meat products were enriched according to ISO 10272-1:2017 procedure A, and the EMRT-PCR in parallel. After 24 h, 12 and 11 samples already tested positive for Campylobacter with the ISO method and EMRT-PCR, respectively. After 40 h, the 24-h-negative sample was also positive with EMRT-PCR. The EMRT-PCR takes about 2 days to produce reliable results, while results using ISO 10272-1:2017 can take up to 8 days, which demonstrate the potential of the EMRT-PCR method.

Role of substrate availability in the growth of Campylobacter co-cultured with extended spectrum beta-lactamase-producing Escherichia coli in Bolton broth.

It is well-established that Extended-spectrum beta-lactamase-producing (ESBL-) Escherichia coli challenge reliable detection of campylobacters during enrichment in Bolton broth (BB) following ISO 10272-1:2017. The overgrowth of Campylobacter by ESBL-E. coli in the enrichment medium BB can lead to false-negative detection outcomes, but the cause for the growth suppression is yet unknown. A plausible reason could be the competition-induced lack of certain growth substrates. Therefore, this study aimed to investigate whether campylobacters and ESBL-E. coli compete for the same medium components and whether this is the cause for the observed growth repression. The availability of possible growth substrates in BB was determined and changes in their extracellular concentration were measured over time during mono-culture enrichment of C. jejuni, C. coli or ESBL-E. coli as well as in co-culture enrichments of campylobacters and ESBL-E. coli. Comparative analysis showed lactate and fumarate utilization by C. jejuni and C. coli exclusively, whereas ESBL-E. coli rapidly consumed asparagine, glutamine/arginine, lysine, threonine, tryptophan, pyruvate, glycerol, cellobiose, and glucose. Both campylobacters and ESBL-E. coli utilized aspartate, serine, formate, a-ketoglutarate and malate. Trends in compound utilization were similar for C. jejuni and C. coli and trends in compound utilization were rather comparable during enrichment of reference and freeze-stressed campylobacters. Since final cell densities of C. jejuni and C. coli in co-cultures were not enhanced by the addition of surplus l-serine and final cell densities were similar in fresh and spent medium, growth suppression seems not to be caused by a lack of substrates or production of inhibitory compounds. We hypothesized that oxygen availability was limiting growth in co-cultures. Higher oxygen availability increased the competitive fitness of C. jejuni 81-176 in co-culture with ESBL-E. coli in duplicate experiments, as cell concentrations in stationary phase were similar to those without competition. This could indicate the critical role of oxygen availability during the growth of Campylobacter and offers potential for further improvement of Campylobacter spp. enrichment efficacy.

A model to predict the fate of Listeria monocytogenes in different cheese types - A major role for undissociated lactic acid in addition to pH, water activity, and temperature.

Undissociated lactic acid has been shown to play a major role in complete growth inhibition of Listeria monocytogenes in Gouda cheese. In addition, low water activity conditions may contribute to growth inhibition. In the current study, it was assessed whether the major factors that inhibit growth of L. monocytogenes in Gouda cheese are the factors that determine growth in other types of ready-to-eat cheese as well. Various types of cheeses were selected, some of which had been associated with listeriosis, while others had not. Based on the composition of the different cheese types, the concentrations of undissociated lactic acid were calculated for each type. The ability to support growth of L. monocytogenes was predicted using the Gamma model, based on literature data on total lactic acid content, moisture content, fat content, pH, Aw, and temperature, and optimal growth rates in milk at 30-37 °C. In addition, the actual specific growth rates of L. monocytogenes in the various cheeses were calculated based on available experimental growth data. In 9 out of the 10 RTE cheeses reviewed, the undissociated lactic acid concentrations and aw determined growth/no growth of L. monocytogenes. No growth was correctly predicted for feta, Cheddar and Gouda, and growth was correctly predicted for ricotta, queso fresco, Camembert, high-moisture mozzarella, cottage and blue cheese. Growth of L. monocytogenes was not observed in practice upon inoculation of Emmental, whereas growth in this cheese type was predicted when including the above mentioned factors in the models. Other factors, presumably acetic and propionic acid, are thought to be important to inhibit growth of the pathogen in Emmental. The results from our study show that for cheeses in which lactic acid is a main acid, our model based on undissociated lactic acid, temperature, pH and aw gives a good prediction of potential outgrowth of L. monocytogenes. Implications for L. monocytogenes legislation are discussed per type of RTE cheese reviewed.

Variability in lag-duration of Campylobacter spp. during enrichment after cold and oxidative stress and its impact on growth kinetics and reliable detection.

Campylobacter jejuni and Campylobacter coli continue to be the leading cause of zoonotic gastroenteritis in the European Union, making reliable detection in food important. Low storage temperatures and atmospheric oxygen concentrations during food production can cause sub-lethal damage or transient non-culturability which is why ISO 10272-1:2017 includes an enrichment step to repair cell damage and increase cell concentrations, thereby supporting detection of campylobacters from foods. The aim of this study was to assess the variability in lag-duration of C. jejuni and C. coli during enrichment after different food-relevant stress treatments and evaluate its impact on growth kinetics and reliability of detection outcomes. Therefore, 13 C. jejuni and 10 C. coli strains were subjected to cold stress during refrigerated and frozen storage. Refrigerated storage did not significantly reduce culturability, but frozen storage reduced cell concentrations by 1.6 ± 0.1 log10cfu/ml for both species. Subsequently, cells were enriched following ISO 10272-1:2017-A and cell concentrations were determined over time and lag-duration and growth rate were determined by fitting the Baranyi-model. Without prior stress treatment, mean lag-duration for C. jejuni and C. coli was 2.5 ± 0.2 h and 2.2 ± 0.3 h, respectively. Refrigerated storage increased lag-duration for C. jejuni to 4.6 ± 0.4 h and for C. coli to 5.0 ± 0.4 h and frozen storage increased lag-duration to 5.0 ± 0.3 h and 6.1 ± 0.4 h for C. jejuni and C. coli, respectively. Comparison of strain- and biological variability showed that differences in recovery after cold stress can be attributed mainly to strain variability since strain variability after refrigeration and freeze stress increased respectively 3-fold and 4-fold while biological variability remained constant. A subset of strains was also subjected to oxidative stress that reduced cell concentrations by 0.7 ± 0.2 log10 cfu/ml and comparison of recovery patterns after oxidative and freeze stress indicated that recovery behaviour was also dependent on the stress applied. A scenario analysis was conducted to evaluate the impact of heterogeneity in outgrowth kinetics of single cells on the reliability of detection outcomes following ISO protocol 10272-1:2017. This revealed that a 'worst-case'-scenario for successful detection by a combination of the longest lag-duration of 7.6 h and lowest growth rate of 0.47 h-1 still resulted in positive detection outcomes since the detection limit was reached within 32.5 h. This suggests that other factors such as competitive microbiota can act as a causative factor in false-negative outcomes of tested food samples.

Effectiveness of a peracetic acid solution on Escherichia coli reduction during fresh-cut lettuce processing at the laboratory and industrial scales.

Fresh leafy greens like lettuce can be consumed raw and are susceptible to foodborne pathogens if they become contaminated. Recently, the number of reported pathogenic foodborne outbreaks related to leafy greens has increased. Therefore, it is important to try to alleviate the human health burden associated with these outbreaks. Processing of fresh-cut lettuce, including washing, is a step in the supply chain that needs to be well controlled to avoid cross-contamination. Current measures to control the quality of lettuce during washing include the use of chemicals like chlorine; however, questions regarding the safety of chlorine have prompted research for alternative solutions with peracetic acid (PAA). This study evaluates the effectiveness of a PAA (c.a. 75 mg/L) solution on the reduction of a commensal E. coli strain during the washing of fresh-cut lettuce. Experiments were performed at the laboratory scale and validated at the industrial scale. We observed that the use of PAA was not adversely affected by the organic load in the water. The contact time and dose of the PAA showed to be relevant factors, as observed by the approximately 5-log reduction of E. coli in the water. Results showed that once introduced during washing, E. coli remained attached to the lettuce, thus supporting the need to control for pathogenic bacteria earlier in the supply chain (e.g., during primary production) as well as during washing. Moreover, our results showed that the use of PAA during washing did not have an apparent effect on the levels of fluorescent pseudomonads (FP) and total heterotrophic bacteria (THB) in lettuce. Overall, our results at the laboratory and industrial scales confirmed that during the processing of fresh-cut produce, where the accumulation of soil, debris, and other plant exudates can negatively affect washing, the use of a PAA (c.a. 75 mg/L) solution was an effective and safe wash water disinfectant that can potentially be used at the industrial scale.

The effect of different matrices on the growth kinetics and heat resistance of Listeria monocytogenes and Lactobacillus plantarum.

Microbial growth and inactivation kinetics in food can be predicted when the effects of food properties and environmental conditions on microbial responses are available. However the effects of these intrinsic and extrinsic variables on microbial kinetics are often obtained using laboratory media, and deviations between predictions and true behaviour might occur if the specific effect of a food product is not known or considered in the prediction. Therefore, knowing the food specific effect on microbial kinetics might not only result in a more realistic growth and inactivation prediction, but also extend the knowledge on factors influencing growth and heat resistance. In this study, growth predictions of Listeria monocytogenes and Lactobacillus plantarum were validated in laboratory media and in milk and ham as model food products. A good agreement between the predicted and observed growth kinetics in laboratory media highlighted the possibility to predict µmax based on cardinal growth parameters obtained from OD-based measurement in laboratory media. Only in two conditions (BHI pH5.5 at 7°C; and BHI pH5.5, undissociated lactic acid concentration of 1mM at 7°C) a possible interaction between growth limiting factors was observed, yet existing interaction models were not better in predicting growth. Growth validation in the two model foods showed that the food specific effects were strain dependent, which might further complicate accurate prediction. For both species the effect of strain variability on thermal inactivation was similar to the food specific effects, and the latter was mainly determined by the effect of ham as heating medium. The combination of both effects explained (almost) all variability found in literature, however, with some bias.

Determination of single cell lag times of Cronobacter spp. strains exposed to different stress conditions: Impact on detection.

The variability of stress resistance and lag time of single cells can have a big impact on their growth and therefore on the probability of their detection in food. In this study, six strains of Cronobacter spp. were subjected to heat, acid and desiccation stress and single cell lag times were determined using optical density measurements. The duration of lag time was highest after acid stress and did not correlate to stress resistance. The effect that the inactivation caused by stress and an extended lag time had on the projected cfu level reached after enrichment was simulated in different scenarios. For most strains, an enrichment time of 18h was sufficient for stressed cells to reach the suggested minimum level of cell inoculum for the Cronobacter screening broth detection. Particular strains may require longer recovery periods. Further, probability calculations showed that the number of samples taken from a batch may have an important effect on detection probability, especially at low contamination rates. Therefore, in addition to increasing the recovery period, increasing the number of samples is a suitable strategy to improve detection.

Quantifying Variability in Growth and Thermal Inactivation Kinetics of Lactobacillus plantarum.

UNLABELLED: The presence and growth of spoilage organisms in food might affect the shelf life. In this study, the effects of experimental, reproduction, and strain variabilities were quantified with respect to growth and thermal inactivation using 20 Lactobacillus plantarum strains. Also, the effect of growth history on thermal resistance was quantified. The strain variability in µmax was similar (P > 0.05) to reproduction variability as a function of pH, aw, and temperature, while being around half of the reproduction variability (P < 0.05) as a function of undissociated lactic acid concentration [HLa]. The cardinal growth parameters were estimated for the L. plantarum strains, and the pHmin was between 3.2 and 3.5, the aw,min was between 0.936 and 0.953, the [HLamax], at pH 4.5, was between 29 and 38 mM, and the Tmin was between 3.4 and 8.3°C. The average D values ranged from 0.80 min to 19 min at 55°C, 0.22 to 3.9 min at 58°C, 3.1 to 45 s at 60°C, and 1.8 to 19 s at 63°C. In contrast to growth, the strain variability in thermal resistance was on average six times higher than the reproduction variability and more than ten times higher than the experimental variability. The strain variability was also 1.8 times higher (P < 0.05) than the effect of growth history. The combined effects of strain variability and growth history on D value explained all of the variability as found in the literature, although with bias. Based on an illustrative milk-processing chain, strain variability caused ∼2-log10 differences in growth between the most and least robust strains and >10-log10 differences after thermal treatment. IMPORTANCE: Accurate control and realistic prediction of shelf life is complicated by the natural diversity among microbial strains, and limited information on microbiological variability is available for spoilage microorganisms. Therefore, the objectives of the present study were to quantify strain variability, reproduction (biological) variability, and experimental variability with respect to the growth and thermal inactivation kinetics of Lactobacillus plantarum and to quantify the variability in thermal resistance attributed to growth history. The quantitative knowledge obtained on experimental, reproduction, and strain variabilities can be used to improve experimental designs and to adequately select strains for challenge growth and inactivation tests. Moreover, the integration of strain variability in prediction of microbial growth and inactivation kinetics will result in more realistic predictions of L. plantarum dynamics along the food production chain.

Effects of different media on the enrichment of low numbers of Shiga toxin-producing Escherichia coli in mung bean sprouts and on the development of the sprout microbiome.

Sprouted seeds have been implicated in a number of serious outbreaks caused by Salmonella and Shiga toxin-producing Escherichia coli. Sprouts pose a very complex challenge to bacterial pathogen enrichment and detection since they naturally contain high levels of background microflora including members of the Enterobacteriaceae. As such, the currently used method cannot ensure reliable detection of STEC in sprouts. In this study, we compared different media for the enrichment of Enterobacteriaceae in their ability to promote the growth of stressed STEC at 37°C and 42°C. Mung bean sprouts were spiked with low levels of STEC and their growth was recorded over time. In addition, the microbiome of mung bean sprouts was analysed before and after enrichment. Our results indicate that the growth of dry-stressed STEC is comparable in all of the tested enrichment media except for mTSB+Novobiocin and not influenced by the incubation temperature. Low levels of STEC spiked into the sprouts resuspended in media only grew to levels of around 4logcfu/ml during enrichment, which could reduce the probability of detection. Proteobacteria was the dominant phylum detected within the microbiome of non-enriched mung bean sprouts. During enrichment in EE-broth, Proteobacteria remained the most abundant phylum. In contrast, during enrichment in BPW the relative abundance of Proteobacteria decreased whereas Firmicutes increased when compared to the non-enriched mung bean sprout microbiome. The microbiome composition was not significantly influenced by the incubation temperature during enrichment in both BPW and EE-broth. This is the first study to examine the microbiome on sprouted mung bean seeds during BPW and EE enrichment and relates the bacterial community composition changes to the enrichment of pathogens.

How NaCl and water content determine water activity during ripening of Gouda cheese, and the predicted effect on inhibition of Listeria monocytogenes.

This study describes the diffusion of NaCl and water in Gouda cheese during brining and ripening. Furthermore, we established water activity as a function of the NaCl-in-moisture content in Gouda cheese during ripening. We determined NaCl content, water content, and water activity in block-type Gouda cheeses that were brined for 3.8d and foil-ripened for a period of 26 wk, and in wheel-type Gouda cheeses that were brined for 0.33, 2.1, or 8.9d and subsequently nature-ripened for a period of 26 wk. The calculated diffusion coefficients of NaCl during brining were 3.6·10(-10) m(2)s(-1) in the block-type Gouda cheeses and 3.5·10(-10) m(2)s(-1) in the wheel-type Gouda cheeses. Immediately after brining, gradients of NaCl and water were observed throughout both types of cheese. During ripening, these gradients disappeared, except for the water gradient in nature-ripened cheeses. An empirical model was derived for Gouda cheese, in which water activity is expressed as a function of the NaCl-in-moisture content, as established for different brining times, locations and ripening times. Moreover, the effect of reduced water activity on inhibition of growth of Listeria monocytogenes in Gouda cheese was calculated. In addition to the presence of lactate and a pH of 5.2 to 5.3, the reduced water activity as seen in Gouda cheese can substantially contribute to inhibition of microbial growth and even to inactivation when cheeses are brined and ripened for extended times and subjected to nature-ripening.

Impact of Pathogen Population Heterogeneity and Stress-Resistant Variants on Food Safety.

This review elucidates the state-of-the-art knowledge about pathogen population heterogeneity and describes the genotypic and phenotypic analyses of persister subpopulations and stress-resistant variants. The molecular mechanisms underlying the generation of persister phenotypes and genetic variants are identified. Zooming in on Listeria monocytogenes, a comparative whole-genome sequence analysis of wild types and variants that enabled the identification of mutations in variants obtained after a single exposure to lethal food-relevant stresses is described. Genotypic and phenotypic features are compared to those for persistent strains isolated from food processing environments. Inactivation kinetics, models used for fitting, and the concept of kinetic modeling-based schemes for detection of variants are presented. Furthermore, robustness and fitness parameters of L. monocytogenes wild type and variants are used to model their performance in food chains. Finally, the impact of stress-resistant variants and persistence in food processing environments on food safety is discussed.

Risk Assessment or Assessment of Risk? Developing an Evidence-Based Approach for Primary Producers of Leafy Vegetables To Assess and Manage Microbial Risks.

Over the last 10 years, some high-profile foodborne illness outbreaks have been linked to the consumption of leafy greens. Growers are required to complete microbiological risk assessments (RAs) for the production of leafy crops supplied either to retail or for further processing. These RAs are based primarily on qualitative judgements of hazard and risks at various stages in the production process but lack many of the steps defined for quantitative microbiological RAs by the Codex Alimentarius Commission. This article is based on the discussions of an industry expert group and proposes a grower RA approach based on a structured qualitative assessment, which requires all decisions to be based on evidence and a framework for describing the decision process that can be challenged and defended within the supply chain. In addition, this article highlights the need for evidence to be more easily available and accessible to primary producers and identifies the need to develop hygiene criteria to aid validation of proposed interventions.

Evaluation of different buffered peptone water (BPW) based enrichment broths for detection of Gram-negative foodborne pathogens from various food matrices.

This study evaluated the effects of changing the composition of the pre-enrichment medium buffered peptone water (BPW) on the growth of stressed and unstressed Gram-negative foodborne pathogens in a one-broth enrichment strategy. BPW supplemented with an available iron source and sodium pyruvate, along with low levels of 8-hydroxyquinoline and sodium deoxycholate (BPW-S) improved the recovery of desiccated Cronobacter spp. from powdered infant formula. Growth of Salmonella and STEC was comparable in all BPW variants tested for different food matrices. In products with high levels of Gram-negative background flora (e.g. sprouts), the target organisms could not be reliably detected by PCR in any of the BPW variants tested unless the initial level exceeded 10(3) cfu/10 g of sprouts. Based on these results we suggest BPW-S for a one-broth enrichment strategy of stressed Gram-negative foodborne pathogens from dry products. However, a one-broth enrichment strategy based on BPW variants tested in this evaluation is not recommended for produce with a high level of Gram-negative background flora due to very high detection limits.

Strain diversity and phage resistance in complex dairy starter cultures.

The compositional stability of the complex Gouda cheese starter culture Ur is thought to be influenced by diversity in phage resistance of highly related strains that co-exist together with bacteriophages. To analyze the role of bacteriophages in maintaining culture diversity at the level of genetic lineages, simple blends of Lactococcus lactis strains were made and subsequently propagated for 152 generations in the absence and presence of selected bacteriophages. We first screened 102 single-colony isolates (strains) from the complex cheese starter for resistance to bacteriophages isolated from this starter. The collection of isolates represents all lactococcal genetic lineages present in the culture. Large differences were found in bacteriophage resistance among strains belonging to the same genetic lineage and among strains from different lineages. The blends of strains were designed such that 3 genetic lineages were represented by strains with different levels of phage resistance. The relative abundance of the lineages in blends with phages was not stable throughout propagation, leading to continuous changes in composition up to 152 generations. The individual resistance of strains to phage predation was confirmed as one of the factors influencing starter culture diversity. Furthermore, loss of proteolytic activity of initially proteolytic strains was found. Reconstituted blends with only 4 strains with a variable degree of phage resistance showed complex behavior during prolonged propagation.

Quantifying strain variability in modeling growth of Listeria monocytogenes.

Prediction of microbial growth kinetics can differ from the actual behavior of the target microorganisms. In the present study, the impact of strain variability on maximum specific growth rate (µmax) (h(-1)) was quantified using twenty Listeria monocytogenes strains. The µmax was determined as function of four different variables, namely pH, water activity (aw)/NaCl concentration [NaCl], undissociated lactic acid concentration ([HA]), and temperature (T). The strain variability was compared to biological and experimental variabilities to determine their importance. The experiment was done in duplicate at the same time to quantify experimental variability and reproduced at least twice on different experimental days to quantify biological (reproduction) variability. For all variables, experimental variability was clearly lower than biological variability and strain variability; and remarkably, biological variability was similar to strain variability. Strain variability in cardinal growth parameters, namely pHmin, [NaCl]max, [HA]max, and Tmin was further investigated by fitting secondary growth models to the µmax data, including a modified secondary pH model. The fitting results showed that L. monocytogenes had an average pHmin of 4.5 (5-95% prediction interval (PI) 4.4-4.7), [NaCl]max of 2.0mM (PI 1.8-2.1), [HA]max of 5.1mM (PI 4.2-5.9), and Tmin of -2.2°C (PI (-3.3)-(-1.1)). The strain variability in cardinal growth parameters was benchmarked to available literature data, showing that the effect of strain variability explained around 1/3 or less of the variability found in literature. The cardinal growth parameters and their prediction intervals were used as input to illustrate the effect of strain variability on the growth of L. monocytogenes in food products with various characteristics, resulting in 2-4 logCFU/ml(g) difference in growth prediction between the most and least robust strains, depending on the type of food product. This underlined the importance to obtain quantitative knowledge on variability factors to realistically predict the microbial growth kinetics.

Reducing viral contamination from finger pads: handwashing is more effective than alcohol-based hand disinfectants.

BACKGROUND: Hand hygiene is important for interrupting transmission of viruses through hands. Effectiveness of alcohol-based hand disinfectant has been shown for bacteria but their effectiveness in reducing transmission of viruses is ambiguous. AIM: To test efficacy of alcohol hand disinfectant against human enteric and respiratory viruses and to compare efficacy of an alcohol-based hand disinfectant and handwashing with soap and water against norovirus. METHODS: Efficacies of a propanol and an ethanol-based hand disinfectant against human enteric and respiratory viruses were tested in carrier tests. Efficacy of an alcohol-based hand disinfectant and handwashing with soap and water against noroviruses GI.4, GII.4, and MNV1 were tested using finger pad tests. FINDINGS: The alcohol-based hand disinfectant reduced the infectivity of rotavirus and influenza A virus completely within 30s whereas poliovirus Sabin 1, adenovirus type 5, parechovirus 1, and MNV1 infectivity were reduced <3 log10 within 3 min. MNV1 infectivity reduction by washing hands with soap and water for 30s (>3.0 ± 0.4 log10) was significantly higher than treating hands with alcohol (2.8 ± 1.5 log10). Washing with soap and water for 30s removed genomic copies of MNV1 (>5 log10), noroviruses GI.4 (>6 log10), and GII.4 (4 log10) completely from all finger pads. Treating hands with propanol-based hand disinfectant showed little or no reduction to complete reduction with mean genomic copy reduction of noroviruses GI.4, GII.4, and MNV1 being >2.6, >3.3, and >1.2 log10 polymerase chain reaction units respectively. CONCLUSIONS: Washing hands with soap and water is better than using alcohol-based hand disinfectants in removing noroviruses from hands.

Risk assessment and risk management for safe foods: Assessment needs inclusion of variability and uncertainty, management needs discrete decisions.

The introduction of relevant food safety changes in legislation, like time-temperature criteria for pasteurisation and sterilisation, microbiological criteria, HACCP and FSOs, generally took several decades. All these approaches have helped to define specific targets or systems to improve the management of food safety. More and more the measures could be related to specific efficiency in public health protection. With the use of quantitative risk assessment, theoretically the effect of all interventions on the final risk can be determined, which can help to design the appropriate controls in the food safety management system. In such an assessment in practice, however results have understandably large variability and also uncertainty. There is large variability and uncertainty in the biological parts of the assessment, the dose response (infectivity, human susceptibility) the micro-organism kinetics in the chain (growth, inactivation, stress response) and also in the more technological parts, the conditions in the chain and the consumer behaviour. Often the results of risk assessments are probability distributions of the variability in illness probability, also sometimes represented with their uncertainty. To make a link from these distributions to managerial decisions, that need to be black and white, should not be considered the job of risk managers. This link needs investment from both the assessor and the manager.

Statistical aspects of food safety sampling.

In food safety management, sampling is an important tool for verifying control. Sampling by nature is a stochastic process. However, uncertainty regarding results is made even greater by the uneven distribution of microorganisms in a batch of food. This article reviews statistical aspects of sampling and describes the impact of distributions on the sampling results. Five different batch contamination scenarios are illustrated: a homogeneous batch, a heterogeneous batch with high- or low-level contamination, and a batch with localized high- or low-level contamination. These batch contamination scenarios showed that sampling results have to be interpreted carefully, especially when heterogeneous and localized contamination in food products is expected.

Quantifying variability on thermal resistance of Listeria monocytogenes.

Knowledge of the impact of strain variability and growth history on thermal resistance is needed to provide a realistic prediction and an adequate design of thermal treatments. In the present study, apart from quantifying strain variability on thermal resistance of Listeria monocytogenes, also biological variability and experimental variability were determined to prioritize their importance. Experimental variability was defined as the repeatability of parallel experimental replicates and biological variability was defined as the reproducibility of biologically independent reproductions. Furthermore, the effect of growth history was quantified. The thermal inactivation curves of 20 L. monocytogenes strains were fitted using the modified Weibull model, resulting in total 360 D-value estimates. The D-value ranged from 9 to 30 min at 55 °C; from 0.6 to 4 min at 60 °C; and from 0.08 to 0.6 min at 65 °C. The estimated z-values of all strains ranged from 4.4 to 5.7 °C. The strain variability was ten times higher than the experimental variability and four times higher than the biological variability. Furthermore, the effect of growth history on thermal resistance variability was not significantly different from that of strain variability and was mainly determined by the growth phase.

Risk assessment strategies as a tool in the application of the Appropriate Level of Protection (ALOP) and Food Safety Objective (FSO) by risk managers.

In the course of the last decade, the Appropriate Level of Protection (ALOP), the Food Safety Objective (FSO) and their associated metrics have been proposed by the World Trade Organization and Codex Alimentarius as a means for competent authorities to ultimately translate governmental public health policy regarding food safety into risk-based targets for the food industry. The industry needs to meet these targets through the effective choice of control measures that are part of its operational food safety management system. The aim of this study was to put the practical application of ALOP and FSO to the test in the case of Salmonella in chicken meat in the Netherlands. Two different risk assessment approaches were applied to derive potential ALOP and FSO values, a 'top-down' approach based on epidemiological data and a 'bottom-up' approach based on food supply chain data. To this end, two stochastic models specific to the Dutch situation were built. Comparisons between 23 countries in Europe were also made using the top-down model. The mean estimated current Level Of Protection values were similar for the two approaches applied, with the bottom-up model yielding 87 cases per 100,000 inhabitants per year (95% CI: 0.03, 904) and the top-down model 71 (95% CI: 9.9, 155). The estimated FSO values on the other hand were considerably different with the mean 'top down' FSO being -4.6 log CFU/g (95% CI: -5.4, -4.1) and the mean 'bottom-up' FSO -6.0 log CFU/g (95% CI: -8.1, -2.9) reflecting major differences in the output distributions of this parameter obtained with the two approaches. Significant differences were observed between current LOP values for different EU countries, although it was not clear whether this was due to actual differences in the factors influencing the risk of salmonellosis or due to the quality of the available data.