Altmetrics
Downloads
112
Views
84
Comments
0
A peer-reviewed article of this preprint also exists.
supplementary.xlsx (11.52KB )
This version is not peer-reviewed
Submitted:
31 January 2024
Posted:
01 February 2024
You are already at the latest version
Scope | Food | RTE | Cross-conta-mination | DR – End-point | Type of DR model | DR Sub-populations | Strain variability | Temp profiles/ Lagtime |
Country | Source |
---|---|---|---|---|---|---|---|---|---|---|
End Processing-to-table | Cold smoked salmon | Yes | No | Exp – I | Pouillot et al. [9] | Multiple | NA | Yes/No | France | Pouillot et al. [8,9] |
Cold smoked salmon | Yes | No | Exp – I | Fritsch et al. [10]: r values from Pouillot et al. [20] were re-scaled to three diff-erent groups of virulence (according to CCs) | General | Specific prevalence for each LM genotypic sub-group (CC) in Europe; Two different distributions for Tmin to represent “low-growing” and “high-growing” strains; Three virulence levels in the DR r values |
Yes/No | France | Fritsch et al. [10] (model based on Pouillot et al. [8,9] integra-ting genomic data) | |
Cold smoked salmon | Yes | No | None | NA | NA | Variable proportion of contaminated packages and growth kinetics parameters according to LM serotypes 1/2a, 1/2b and 4b | No/Yes | USA | Chen et al. [11] | |
Retail-to-table | Various: smoked seafood, raw seafood, preserved fish, cooked RTE crustaceans | Yes | No | Mouse Epi – I | FDA-FSIS [5] | Multiple | Variability in the virulence of different strains represented in DR | No/No | USA | FDA-FSIS [5] |
Packaged cold-/hot-smoked fish and gravad fish | Yes | No | Exp – I | Pouillot et al. [20] | Multiple | Challenge test data from a mixture of strains; h0 distribution of variability in physiological state of cells; variability in strain virulence and in susceptibility across population subgroups | Yes/Yes | EU | Pérez-Rodríguez et al. [12] | |
Cold-, hot-smoked fish, gravad fish | Yes | No | Exp – I | EFSA BIOHAZ [1] based on Pouillot et al. [20] | Multiple (sex/age group) | Challenge test data from a mixture of strains; Strain virulence and host susceptibility explicit in r distribution | No/No | EU | EFSA BIOHAZ [1] | |
Consump-tion | Smoked/gravad salmon/rainbow trout | Yes | No | Exp – I | Buchanan et al. [21] | General | All strains are virulent vs a proportion of virulent strains | No/No | Sweden | Lindqvist and Westöö [13] |
Cold smoked fish | Yes | No | Exp – I | FAO-WHO [14] | High-risk/Low-risk | Strain diversity implicit in r | No/Yes | Non-specific | FAO-WHO [14] | |
Smoked fish and sliced cooked ham | Yes | No | Exp – I | FAO-WHO [14] | High-risk/Low-risk | Strain diversity implicit in r | No/No | Spain | Garrido et al. [15] | |
Cold smoked salmon | Yes | No | BP – I | Haas et al. [22] | General | NA | No/Yes | Non-specific | Gospavic et al. [16] | |
VP cold smoked salmon | Yes | No | WG – I | Farber et al. [23] | High-risk/Low-risk | Challenge test data from a mixture of strains; | No/Yes | Ireland | Dass [17] | |
Traditional processed fish | No | No | WG – I | Farber et al. [23] | High-risk/Low-risk | NA | No/No | Ghana | Bomfeh [18] | |
Cold smoked and salt-cured fishery products | Yes | No | Exp – I | Pasonen et al. [19] | High-risk/Low-risk | NA | No/No | Finland | Pasonen et al. [19] |
Scope | Food | Predictive microbiology models | What-if scenarios | Sensitivity analysis | Model complexity | Source |
---|---|---|---|---|---|---|
End processing-to-table | Cold smoked salmon | Growth (Jameson effect LM and background microflora, growth square root models for LM and background microflora) |
EXPOSURE ASSESSMENT: (1) Reducing theoretical shelf-life from 28 days to 15 days reduced mean LM/g in contaminated servings in 10%; (2) The baseline scenario of 21.4% of shelf lives at home longer than 7 days was compared to a scenario of consumption within 7 days maximum; and reduces the mean LM/g in 10%; (3) Better refrigeration at retail from mean temperature 5.6 to 4ºC reduces the mean LM counts in 19%; (4) Better refrigeration at home from mean temperature 7 to 4ºC, reduces the mean LM counts in 36%; (5) Lower initial concentration from 0.46% of values above 1 CFU/g to a distribution truncated at 1 CFU/g, reduces the mean LM counts in 8%. RISK ASSESSMENT: Output – Listeriosis cases compared to a base 100 for the baseline model: (1) Shelf-life 15 days =23; (2) Prevalence of LM to a quarter=25; (3) Mean home refrigerator temperature 4ºC = 34; (4) Consumed 7 days after purchase = 37; (5) Prevalence of LM to a half = 50; (6) Mean retail temperature at 4ºC=67. |
EXPOSURE ASSESSMENT: Output – concentration of LM in contaminated servings: (1) Total duration at the consumer phase (p=10-30), (2) Mean temperature at the consumer phase (p=10-20), (3) Initial LM counts (p=10-20), (4) Mean temperature at retail phase (p=10-14), (5) Total duration of retail phase (p=10-8), (6) Tmin for growth (p=10-8), (7) Tmin microflora (p=10-6), (8) Initial background flora counts (p=0.002), (9) Serving size (p=0.003), (10) MPD (p=0.008), (11) Ref GR at 25ºC (p=0.015), (12) Ref GR of flora at 25º C (p=0.025). RISK ASSESSMENT: Output - listeriosis cases in the reference population: (1) r value of DR model (p=10-300), (2) SD(MPD) (p=10-137), (3) Ref of GR 25ºC for LM (p=10-101), (4) MPD of LM (p=10-76), (5) Tmin of LM (p=10-12), (6) GR of flora 25ºC (p=10-8), (7) Prevalence of LM (p=10-6), (8) Servings/year (p=10-2). |
Medium: Complex predictive microbiology model; a new method for solving growth under dynamic temperature profiles was proposed. | Pouillot et al. [8,9] |
Cold smoked salmon | Growth (Jameson effect LM and background microflora, growth square root models for LM and background microflora) | Baseline predicted 978 listeriosis cases after consumption of 50 g cold smoked salmon with an initial LM prevalence of 10.4% considering one single prevalence distribution. (1) Taking into account specific prevalences for each LM genotypic sub-group lowered the cases to 574 listeriosis cases; (2) 97% of listeriosis cases were caused by the hypervirulent group despite their low prevalence (12.6%) in contaminated salmon. Inversely, the most prevalent (hypovirulent) group 51.7%) was responsible for only 0.02% of the listeriosis cases; (3) The effect of the low/high growth strains (two distributions for Tmin) was lower than the effect of the virulence: mean exposure from high growth LM group was 25 CFU/g, compared to the low growth groups (13 CFU/g). | NA | Medium: Same as Pouillot et al. [8,9] but with the further complexity of adding phenotypic characteristics of LM by subgroup and virulence properties of LM. | Fritsch et al. [10] (model based on Pouillot et al. [8,9] integra-ting genomic data) | |
Cold smoked salmon | Growth models (Buchanan, Gompertz and Baranyi as primary models, and secondary square root model); and Die-off and re-growth models (Weibull-Buchanan, Weibull-Gompertz and Weibull-Baranyi) | End point of the model is the regulatory and recall risk (RRR) defined as the overall risk of a lot sampled found positive for LM. (1) Treatment of salmon with 5 or 20 ppm nisin reduced RRR to 0.109 or 0.017 (in comparison to baseline RRR of 0.333); (2) Reducing prevalence to half decreased RRR to 0.182; (3) Use of inhibitors (2% potassium lactate + 0.14% sodium diacetate) slightly reduced RRR to 0.313; (4) Keeping cold storage below 5ºC did not reduce RRR. | Output – regulatory and recall risk: (1) initial contamination level (r=0.404), (2) GR at 25ºC (r=0.275), (3) storage temperature (r=0.177), (4) Tmin (r=-0.169), (5) MPD (r=0.053) | Medium: Uncertainty and variability are separated; the die-off and/or growth kinetics are too compartment-alised. | Chen et al. [11] | |
Retail-to-table | Various: smoked seafood, raw seafood, preserved fish, cooked RTE crustaceans | Growth (linear model, EGR5 square root models) | (1) For cold smoked salmon, reducing the max home storage time from 45 to 30 days, reduces the mean cases in 38% in the elderly population. | NA | Medium: Various foods |
FDA-FSIS [5] |
Packaged cold-/hot-smoked fish and graved fish | Growth (Baranyi model with Jameson effect LM and LAB, EGR5 square root model and effect of lactate) | (1) Decreasing the maximum initial LM concentration by 2 log decreases listeriosis cases per million servings in >99%; (2) Decreasing time to consumption in 25% decreases listeriosis in 80%; (3) Decreasing 1-2 ºC in the dynamic temperature profiles reduces cases in 75%; (4) Including lag time in the model has no effect on listeriosis cases. | NA | Medium: Time temperature dynamic profiles from retail to consumption, and microbial competition models used solved with RK4 algorithm. |
Pérez-Rodríguez et al. [12] | |
Cold-, hot-smoked fish, gravad fish | Growth (Rosso model, EGR 5ºC) | (1)Across the 3 RTE fish products, there is no strong difference in the probability of a product exceeding 100 CFU/g at the time of consumption between normal packaging (0.066 – 0.112) and reduced-oxygen packaging (0.040 – 0.115); (2) In both, reduced-oxigen and normal packaging, hot-smoked fish presented higher probability of exceeding 100 CFU/g at the point of consumption (0.115, 0.112) than cold-smoked fish (0.080, 0.074) and gravad fish (0.047, 0.066). | Risk is very sensitive to MPD. A shift in 0.5 log CFU/g can double the estimated risk. However, sensitivity analysis was conducted taking together various RTE food classes. | Low: Generic model; only demands some knowledge in R software to utilise it | EFSA BIOHAZ [1] | |
Consump-tion | Smoked / gravad salmon / rainbow trout | NA | (1) The minimum level of LM resulting in a risk of illness greater than 10-7 or 10-8 was 25 or 2 CFU/g; (2) If the assumption that all strains are virulent was reduced to 1-10%, the annual listeriosis cases is reduced by 84% in the both high-risk and the low-risk populations. | Output – annual risk of illness: ranked as initial LM counts, prevalence, serving size, proportion of virulent strains | Low | Lindqvist and Westöö [13] |
Cold smoked fish | Growth (LM growth model affected by LAB growth, Square root model for GR as a function of tempera-ture, pH, aw, un-dissociated lactic acid) | (1) If LM growth rate inhibition due to LAB growth is between 80-100%, the increase of listeriosis per 100 000 people is 684-fold in the overall population, in comparison to the baseline scenario of no-growth of LM between purchase and consumption; (2) If LM growth rate inhibition due to LAB growth is 95%, the increase of listeriosis per 100 000 people is 67-fold in the overall population in comparison to the no-growth of LM baseline scenario; (3) Reducing the mean shelf-life of smoked fish from 14 to 7 days, results in an 80% reduction of listeriosis. | NA | Medium: Relative lag time concept for LM and LAB |
FAO-WHO [14] | |
Smoked fish (salmon and trout) | Growth (Logistic model without delay, growth cardinal model) | (1) Reducing home storage time from a max of 30 to 7 days, reduces the annual cases in 15% for salmon and in 45% for trout;(2) If all domestic temperatures had a mean temperature of 4.5ºC, the mean annual cases is reduced in 65% for salmon and in 70% for trout; (3) Combi-ning the two measures above reduces the mean annual cases in 75% for salmon and in 87% for trout; (4) If at purchase, LM counts would not exceed 100 CFU/g (truncating the baseline N~(1.01, 0.71) for smoked salmon and N~(1.35,1.40) for smoked trout, the mean annual cases would decrease in 22% in salmon and in 99% in trout. | NA | Low | Garrido et al. [15] | |
Cold smoked salmon | Growth (Baranyi model with Jameson effect LM and background microflora, Extended GR models for LM and LAB as a function of temperature, pH, aw, un-dissociated lactic acid, undissociated diacetate, phenols, dissolved CO2 and nitrite) | (1) At a mean initial LM counts of 4 CFU/g, reducing the time of consumption from 28 to 14 days reduces the risk of illness in 64%; (2) If mean time of consumption is 14 days, reducing the mean initial counts from 25 CFU/g to 4 CFU/g reduces the risk of illness in 67%. | NA | Medium: stochastic fluctuations in the GR of LM are taken into account by using white noise and the Winner process. |
Gospavic et al. [16] | |
Vacuum-packed cold smoked salmon | Growth (Baranyi model, growth square root model) | (1) If initial LM counts at retail (1-1000) was truncated at >100 CFU/g, the risk of illness would reduce in 0.3/0.9 log (high-risk and low-risk populations); (2) Reducing the maximum consumer shopping time from 3 hours to 30 min reduces risk of illness in 0.8/0.8 log; (3) Reducing consumer storage days from 21-30 to 7-15 days reduces risk of illness in 0.5/0.6 log; (4) Fixing storage temperature from 3-10ºC to 4ºC reduces risk of illness in 1.0/1.1 log; (5) If LM counts were not higher than 2 log CFU/g and reducing the maximum shopping time to 30 min, reducing consumer storage days to 7-15 days and storage temperature to 4ºC, reduces risk of illness in 1.32/1.39 log. | Output – annual risk of illness in the high-risk population: (1) LM counts at retail (r=0.97); (2) Temperature in consumer fridge (r=0.13); (3) Time in consumer fridge (r=0.06). | Low: Lag: Baranyi model with bacterial adaptation |
Dass [17] | |
Traditional processed fish | NA | NA | NA | Low | Bomfeh [18] | |
Cold smoked and salt-cured fishery products | Growth (Logistic growth model, growth cardinal parameter model as a function of temperature, salt content, pH and phenolic compounds) | (1) If home storage temperature decreased from 7ºC to 3ºC, the median cases of listeriosis per 100 000 elderly population would decrease in 70%.; (2) If home storage temperature decreased from 7ºC to 3ºC, the median cases of listeriosis per 100 000 working-age population would decrease in 40%. | NA | High: Parameters, including r, were estimated from a Bayesian model consisting of three linked modules: a model for the occurrence data, a model for the consumption data and a predictive model for the total number of cases in the population. The current model takes into account the possi-bility of continuing consumption of the same (contamina-ted) package of CSS/SCS, rather than assuming independent consumption days. | Pasonen et al. [19] |
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content. |
© 2024 MDPI (Basel, Switzerland) unless otherwise stated