Digging into Toxoplasma gondii infections via soil: A quantitative microbial risk assessment approach.

Soil has been identified as an important source of exposure to a variety of chemical and biological contaminants. Toxoplasma gondii is one of those potential biological contaminants associated with serious health effects in pregnant women and immunocompromised patients. Gardening or consumption of homegrown vegetables may present an important route of T. gondii infection via accidental ingestion of soil. In the Netherlands, there is quantitative information on the risk of T. gondii infection via meat products, but not on the risk of infection through soil. The objective of this study was to develop a quantitative microbial risk assessment (QMRA) model for estimating the risk associated with T. gondii exposure via accidental soil ingestion in the Netherlands. In order to obtain the needed information, a magnetic capture method for detection of T. gondii oocysts in soil samples was developed, and T. gondii DNA was detected using qPCR targeting the 529 bp repeat element. The method was shown to provide 95% probability of detection (95% CI: 88-100%) when at least 34 oocysts are present in 25 g of soil. T. gondii DNA was detected in 5 of 148 soil samples with interpretable results (3%, 95% CI: 1.5-7.7%). Results for 18 samples were not interpretable due to PCR inhibition. The estimated amount of oocysts presented in qPCR positive samples was quantified by a linear model, and the amount varied from 8 to 478 in 25 g of soil. The estimated incidence rate of T. gondii infection from the QMRA model via soil varied from 0.3 to 1.8 per 1000 individuals per day. Several data gaps (e.g., soil contamination/ingestion and oocysts viability) have been identified in this study, the structure of the model can be applied to obtain more accurate estimates of the risk of T. gondii infection via soil when data become available.

The effect of salting on Toxoplasma gondii viability evaluated and implemented in a quantitative risk assessment of meat-borne human infection.

The protozoan parasite Toxoplasma gondii can infect all warm-blooded animals and it causes the disease toxoplasmosis. Meat containing viable T. gondii tissue cysts is considered one of the main sources of human infection. The relative importance of the different types of meat depends, not only on the prevalence of T. gondii infection in the different livestock species, but also on consumed volumes and preparation habits. To take these factors into account and to estimate the relative contribution of different meat products to human infection, a quantitative risk assessment model for meat-borne T. gondii infection was previously developed. However, at the time, the effect of salting on parasite viability was estimated based on a single experiment. In recent years, data using salting methods that are more in line with processing of meat products have come available. Literature data on the effect of salting on T. gondii viability were collected and used to fit a predictive model. In addition to the new salting model, a lower concentration of bradyzoites in cattle, more specific heating profiles, and more recent consumption data were implemented in the QMRA model for meat-borne T. gondii infection in the Netherlands. Results show that beef remains the most important source, as it contributed 84% of the total number of predicted infections in the Dutch population, followed by pork (12%), mutton (3.7%), lamb (0.2%) pork/beef mixed products (0.1%), and veal (0.01%). The predicted number of T. gondii infections is reasonably in line with epidemiological data. At the product level, filet americain (a raw beef spread) alone contributed 80% of the total predicted infections in the base model, but scenario analyses demonstrate that its contribution is highly dependent on the salting parameters. A clear identification of the most risky meat products is important, as interventions focussing on these products could have a great impact on reducing T. gondii disease burden in the Netherlands. For that reason, it is important that the effects of salting and other processing methods are evaluated in line with industrial processing and incorporated in quantitative risk assessment models for meat-borne toxoplasmosis.

A social cost-benefit analysis of two One Health interventions to prevent toxoplasmosis.

In the Netherlands, toxoplasmosis ranks second in disease burden among foodborne pathogens with an estimated health loss of 1,900 Disability Adjusted Life Years and a cost-of-illness estimated at €45 million annually. Therefore, effective and preferably cost-effective preventive interventions are warranted. Freezing meat intended for raw or undercooked consumption and improving biosecurity in pig farms are promising interventions to prevent Toxoplasma gondii infections in humans. Putting these interventions into practice would expectedly reduce the number of infections; however, the net benefits for society are unknown. Stakeholders bearing the costs for these interventions will not necessary coincide with the ones having the benefits. We performed a Social Cost-Benefit Analysis to evaluate the net value of two potential interventions for the Dutch society. We assessed the costs and benefits of the two interventions and compared them with the current practice of education, especially during pregnancy. A 'minimum scenario' and a 'maximum scenario' was assumed, using input parameters with least benefits to society and input parameters with most benefits to society, respectively. For both interventions, we performed different scenario analyses. The freezing meat intervention was far more effective than the biosecurity intervention. Despite high freezing costs, freezing two meat products: steak tartare and mutton leg yielded net social benefits in both the minimum and maximum scenario, ranging from €10.6 million to €31 million for steak tartare and €0.6 million to €1.5 million for mutton leg. The biosecurity intervention would result in net costs in all scenarios ranging from €1 million to €2.5 million, due to high intervention costs and limited benefits. From a public health perspective (i.e. reducing the burden of toxoplasmosis) and the societal perspective (i.e. a net benefit for the Dutch society) freezing steak tartare and leg of mutton is to be considered.

A quantitative microbial risk assessment for meatborne Toxoplasma gondii infection in The Netherlands.

Toxoplasma gondii is an important foodborne pathogen, and the cause of a high disease burden due to congenital toxoplasmosis in The Netherlands. The aim of this study was to quantify the relative contribution of sheep, beef and pork products to human T. gondii infections by Quantitative Microbial Risk Assessment (QMRA). Bradyzoite concentration and portion size data were used to estimate the bradyzoite number in infected unprocessed portions for human consumption. The reduction factors for salting, freezing and heating as estimated based on published experiments in mice, were subsequently used to estimate the bradyzoite number in processed portions. A dose-response relation for T. gondii infection in mice was used to estimate the human probability of infection due to consumption of these originally infected processed portions. By multiplying these probabilities with the prevalence of T. gondii per livestock species and the number of portions consumed per year, the number of infections per year was calculated for the susceptible Dutch population and the subpopulation of susceptible pregnant women. QMRA results predict high numbers of infections per year with beef as the most important source. Although many uncertainties were present in the data and the number of congenital infections predicted by the model was almost twenty times higher than the number estimated based on the incidence in newborns, the usefulness of the advice to thoroughly heat meat is confirmed by our results. Forty percent of all predicted infections is due to the consumption of unheated meat products, and sensitivity analysis indicates that heating temperature has the strongest influence on the predicted number of infections. The results also demonstrate that, even with a low prevalence of infection in cattle, consumption of beef remains an important source of infection. Developing this QMRA model has helped identify important gaps of knowledge and resulted in the following recommendations for future research: collect processing-effect data in line with consumer style processing and acquire product specific heating temperatures, investigate the presence and concentration of viable bradyzoites in cattle, determine the effect of mincing meat on bradyzoite concentrations using actual batch sizes, and obtain an estimate of the fraction of meat that has been frozen prior to purchase. With more accurate data this QMRA model will aid science-based decision-making on intervention strategies to reduce the disease burden from meatborne T. gondii infections in The Netherlands.