Public health impact and control options of parasitic zoonoses via pork

Introduction

Parasites have long been neglected but nowadays are becoming more recognized as important foodborne pathogens. Various factors contribute to their undervaluation such as complex lifecycles and transmission routes, prolonged incubation period and chronic disease manifestations. In addition, rapid and sensitive diagnostic assays are not always available and therefore parasite occurrences are often underreported, resulting in low disease burden estimates, compared to viral or bacterial causes of foodborne diseases (Newell, Koopmans et al. 2010). In 2015, WHO Foodborne Disease Epidemiology Reference Group (FERG) reported the global disease burden of foodborne diseases expressed in disability adjusted life years (DALYs) of 31 infectious pathogens and chemical hazards (Havelaar, Kirk et al. 2015). Among the highest-ranking etiological agents of these foodborne diseases, many originated from animals (zoonotic pathogens) and included not only bacteria such as Salmonella and Campylobacter spp. but also foodborne parasites such as Taenia solium (T. solium) in the top list.

T. solium causing taeniasis and neurocysticercosis, is an important zoonotic pathogen including humans and pigs. It has the highest global disease burden of all parasitic foodborne diseases with 2.8 million DALYs (Havelaar, Kirk et al. 2015). It mainly affects poor communities in low- and middle-income geographic areas, such as in Africa, South America and South East Asia and are indicative of poor hygiene standards of pig husbandry practices. T. solium is on the WHO list of neglected tropical parasites, with the high DALY estimation mainly caused by neurocysticercosis (Torgerson, Devleesschauwer et al. 2015). Therefore, control and elimination programmes should be high priority and goals were set for interventions in selected countries by WHO ((World Health and Thomas 2015)).

In addition, two other parasitic zoonoses were also included in the FERG disease burden estimations. Toxoplasma gondii (T. gondii) causing toxoplasmosis, Trichinella spiralis (T. spiralis) and other Trichinella species causing trichinellosis. Those are both important pathogens that can be transmitted from pigs to humans. Toxoplasma gondii was ranked after T. solium as one of the highest ranked foodborne parasites (Havelaar, Kirk et al. 2015, Torgerson, Devleesschauwer et al. 2015). Trichinellosis had the lowest disease burden estimated by FERG (Devleesschauwer, Praet et al. 2015), however, it is the only foodborne parasite that is mandatory controlled in many countries for which international guidelines for trade are available (http://www.fao.org/fao-whocodexalimentarius/codex-texts/guidelines/en/). Although it remains questionable whether the low DALYs for trichinellosis are the consequence of these control efforts. In addition, the Codex Alimentarius Committee on Food Hygiene (CCHF) (http://www.fao.org/fao-who-codexalimentarius/en/) became more interested in the importance of foodborne parasites and requested FAO and WHO ‘ to review the current status of knowledge on parasites in food and their public health and trade impact in order to provide CCHF with advice and guidance on the parasite-commodity combinations of particular concern’ in order to support CCHF to develop more general guidelines for the control of foodborne parasites. In 2012, experts from five continents were invited to prioritize foodborne parasites (FBP) using multicriteria decision analysis, an approach often used to rank diseases in a way that comparisons can be made. Of the twenty-four foodborne parasites of global concern, T. solium ranked the highest followed by Echinococcus multilocularis, E. granulosus and T. gondii (Robertson, van der Giessen et al. 2013, World Health, Food et al. 2014). One of the main recommendations was that the regional ranking could be very different from the global ranking and the same exercise should be carried out at the regional level). During a European financed COST Action the European network of foodborne parasites (https://www.euro-fbp.org/) , the same methodology was repeated for Europe. In Europe, Echinococcus multilocularis was the highest ranked FBP followed by T. gondii, E. granulosuT. spiralis s and (and other Trichinella spp.) (Bouwknegt, Devleesschauwer et al. 2018). ). In Europe, the impact of T. solium was not clear and studied in another COST action Cystinet (https://www.cost.eu/actions/TD1302).

In order to study the public health impact of these three foodborne and zoonotic parasites, we developed quantitative risk assessment (QMRA) models with the following aims:

1. Toxoplasma gondii: to get more insight in the relative importance of meat products for human infections;

2. Trichinella spiralis: to compare the public health risks of pigs kept under controlled and non-controlled housing conditions;

3. Taenia solium: to assess the risk of human T. solium exposure from home slaughtered pigs versus slaughtered pigs with meat inspection.

In this paper we summarize the framework of the QMRA models per FBP and the results. The results can be used to give directions for control options to reduce the disease burden of these foodborne diseases.

Quantitative microbiological risk assessment for three foodborne parasitic diseases

Toxoplasmosis

The life cycle of T. gondii entails two main infectious stages: 1) oocysts are excreted in the environment by recently infected felines (definitive host) and become infectious to virtually all warm-blooded animals upon sporulation, 2) tissue cysts develop in muscle and nervous tissues of intermediate hosts and are infectious to felines and other intermediate hosts via carnivory. Thus, consumption of raw or undercooked meat is one of the main sources of T. gondii infection in humans. In case a woman is primary infected during pregnancy, the parasite can be transmitted transplacentally and result in abortion, or congenital disease in the newborn. The classical triad for congenital toxoplasmosis consists of chorioretinitis, intracranial calcifications and hydrocephalus, however, often newborns show other less specific symptoms or are asymptomatic at birth. In immunocompetent individuals, the acute phase of the infection usually passes asymptomatically or signs are limited to lymphadenopathy and mild flu-like symptoms. However, both congenitally infected children and immunocompetent individuals that acquired an infection later in life are at risk for developing chorioretinitis. In severely immunocompromised patients (e.g. AIDS-patients or patients under strong immunosuppressive treatment), uncontrolled parasite replication from primary or by recrudescence of a latent infection can result in potentially fatal encephalitis, pneumonitis, myocarditis or disseminated toxoplasmosis. Mainly the lifelong health implications of congenital toxoplasmosis, but also the effects of ocular toxoplasmosis after acquired infection contribute to the disease burden in DALYs (Havelaar, Kirk et al. 2015, Torgerson, Devleesschauwer et al. 2015).

QMRA meat borne Toxoplasma gondii

To study the relative contribution of different meat products a quantitative risk assessment model was set up (Opsteegh, Prickaerts et al. 2011). In this model (Fig. 1) data on the prevalence and concentration of T. gondii in livestock was combined with data on preparation and consumption of various meat products. Moreover, inactivation by meat processing (i.e. heating, freezing and salting) was modelled and a dose-response model based on data from animal experiments (Guo, Mishra et al. 2016) was used to predict the probability of infection per consumed portion. Despite a low prevalence of T. gondii infection in cattle, consumption of beef is predicted to be the main source of T. gondii infection via meat in the Netherlands (Opsteegh, Prickaerts et al. 2011, Deng, Swart et al. 2019). Pork products were estimated to contribute 11-12% of meat borne infections. The relative contribution is strongly influenced by preparation and consumption habits. In the Netherlands, fresh meat is generally heated properly and two popular ready-to-eat raw meat products were most important: “Filet americain” (a raw beef spread) and “theeworst” (a raw pork sausage) add up to 90% of the total of predicted meat borne infections (Deng, Swart et al. 2019).

Figure 1. Schematic overview of the quantitative risk assessment model for meat borne T. gondii infections. (From: (Opsteegh, Prickaerts et al. 2011).

The QMRA model developed for the Netherlands was applied for mainland China, using average consumption data per day per livestock species and preparation habits based on American consumers. Here, pork and chicken were predicted to be the most important sources of meat borne infection (Deng, Swart et al. 2019). However, the predicted number of infections was at least thousand-fold higher than estimated based on epidemiological data, which demonstrates the need for detailed and country-specific data on consumption and preparation habits.

Trichinellosis

Nematodes belonging to Trichinella spp. can cause trichinellosis in humans. T. spiralis has a worldwide distribution and infected pigs are a main reservoir for human infections. Trichinellosis is a serious disease in humans and clinical outcome can vary from mild to very severe. High dose infections can result in acute heart failure and death. Control of Trichinella in pigs by the end of the nineteenth century was the start of meat inspection in Europe. Nowadays, control is still mandatory under EU legislation in Europe and many third countries for trade of pork (van der Giessen, Franssen et al. 2013) (Franssen, van Andel et al. 2016). Pigs kept under controlled housing conditions in Europe showed a negligible risk and most outbreaks occur after consumption of meat originating from backyard pigs and wild boar. Therefore, risk based surveillance and control is an option and quantifying human risk can give direction to risk based control options.

QMRA Trichinellaspiralis

The QMRA-T workflow started from parasite prevalence and abundance for pigs originating from controlled and non-controlled housing. Trichinella distributions between and within pigs were modelled and the outcome was used to estimate the probability to encounter Trichinella larvae at meat inspection. From there, probabilities of encountering Trichinella larvae in pig portions were modelled and subsequently, the probability to become infected from consuming cooked portions of pork was quantified, using an earlier published dose-response model for Trichinella in humans (Teunis, Koningstein et al. 2012). Finally, we estimated human incidence of trichinellosis for the European Union. Moreover, the effect of test sensitivity on human trichinellosis incidence from pigs from non-controlled housing was quantified.

The estimated annual risk from pigs from non-controlled housing was 59,443 human trichinellosis cases without testing at meat inspection and 832 (95%CI 346 – 1410) cases with Trichinella testing, thus preventing 98.6% of trichinellosis cases per year by testing at meat inspection. Using the QMRA-T, already a 5% decrease in test sensitivity resulted in four times higher human trichinellosis cases from this housing type, compared to baseline incidence.

The estimated annual risk for pigs from controlled housing was less than 0.002 (range 0.000 – 0.007) human cases with- and less than 0.010 (0.001 – 0.023) cases without Trichinella testing at meat inspection, which does not differ significantly (p=0.2075). In practice, this means no cases per year irrespective of Trichinella testing. Thus, controlled housing effectively prevents infection and Trichinella testing does not contribute to food safety for this housing type. Not testing for Trichinella requires evidence based full compliance with regulations for controlled housing (Franssen, Takumi et al. 2018).

Figure 2. Schematic overview of the quantitative risk assessment model for meat borne T. spiralis infections. (From: Franssen et al., (2017), International Journal of Food Microbiology 241 262–275).

Schematic overview of the quantitative risk assessment model for meat borne T. spiralis infections. (From: Franssen et al., (2017), International Journal of Food Microbiology 241 262–275).

Taenia solium/cysticercosis

Taenia solium is a zoonotic tapeworm with pigs as the intermediate host and humans as the definitive host. The adult stage of the parasite in the human host can cause taeniasis when raw or undercooked meat of infected pigs is consumed. In humans, the zoonotic tapeworm produces hundreds of eggs, which are released in the environment through faeces. Humans can also act as intermediate hosts when colonized by embryonated eggs, either through autoinfection/self-infection or contaminated food or water (cysticercosis). When cysts localize in the central nervous system, it might lead to neurocysticercosis (Trevisan, Sotiraki et al. 2018).

QMRA Taenia solium/cysticercosis

QMRA T. solium

The aim of this QMRA model was to quantify the risks of human T. solium exposure from consuming pork originating from pigs slaughtered at home and controlled slaughterhouses for five European countries (i.e. Bulgaria, Germany, Poland, Romania and Spain) (Meester, Swart et al. 2019). The model took different stages of the food chain into account and comprised three parts, started from production, then inspection, and consumption. Data on the prevalence of porcine cysticercosis, the percentage of home slaughtered pigs, the sensitivity of meat inspection, cyst distribution in pork, and pork consumption amount were collected and used as inputs for the QMRA model (Fig. 3). The question was if consumption of undercooked meat from home slaughtered pigs without meat inspection is a higher risk for humans than pigs under controlled conditions with meat inspection in Europe.

Figure 3. Conceptual risk chain of Taenia solium exposure (Meester, Swart et al. 2019).

At the production part, the model sets off with the reported porcine cysticercosis prevalence and adjusted it with sensitivity of meat inspection. Then, the rate of pigs exposed to eggs and infection loads of porcine cysticercosis in muscles were calculated. In the end of the production part, the number of infected pigs for both home slaughtered and controlled conditions were determined. Since meat inspection is obligatory at slaughterhouses in European Union, the inspection part was included in the controlled condition. At the consumption part, all the false negative/infected carcasses from the two conditions estimated at previous steps were included. And then, the probability of a cyst to enter specific meat cuts, and cyst distribution in consumed portions were predicted. Finally, after applying four cooking scenarios, the prevalence of infected portion originated from controlled slaughterhouses and home slaughtered pigs were estimated.

Results showed that the prevalence of contaminated pork portions from home slaughtered pigs was almost 14 times higher than from slaughterhouses. This was mainly influenced by the prevalence of cysticercosis in pigs. The sensitivity of T. solium inspection depended on the infection load of the heart, and each heart cysticercus has a probability of 0.32 to be detected. Although the test sensitivity of inspection was low, it did not affect the risk of exposure. The results of cooking scenarios showed that cooking can effectively reduce the risk of exposure to T. solium infected pork (Meester, Swart et al. 2019).

Discussion and conclusion

Quantitative risk assessment can direct risk-based surveillance and control options. In this paper, we have shown QMRA models developed for three different meat transmitted parasitic infections. In case of Toxoplasma gondii, the QMRA models included the prevalence of infection in the main meat producing animal species, concentration of bradyzoites in meat, dose response relationship, meat consumption data and T. gondii inactivation characteristics i.e. salting, heating and freezing of meat. The QMRA results showed that beef was most important and pork products contributed to 11-12% of predicted meat borne infections in the Netherlands (Deng, Swart et al. 2020). The relative contribution was strongly influenced by preparation and consumption habits and can differ from country to country. In the Netherlands, two popular ready-to-eat raw meat products were most important (Deng, Swart et al. 2019). This has guided further research towards the effects of meat processing and may lead to intervention measures targeted at these specific products, either by freezing of the meat used for production (Suijkerbuijk, Over et al. 2019) or by adjusting other processing parameters (e.g. salt concentration) to ensure T. gondii inactivation. In China, food habits are very different from European food habits and here pork and chicken were predicted to be the most important sources of meat borne infection (Deng, Swart et al. 2019), however human consumption data in China were only available at the livestock level. Therefore, to identify the risky meat products contributing most to human infections in other countries, the QMRA model can be used as long as country-specific food consumption and preparation data are collected using a similar method. A project aiming to study meat borne source attribution just started under the European Joint Programme One Health (https://onehealthejp.eu/jrp-toxosources/). The described QMRA model will be assessed with food consumption data from nine European countries to obtain better insight of control options in Europe.

To evaluate control options for Trichinella infections in pigs, a QMRA was developed starting from the risk of wild boar, backyard pigs and pigs under controlled housing conditions. The QMRA model predicted the number of infections in Europe from backyard pigs and controlled housing pigs and results showed that meat inspection for Trichinella reduced the number of infections after the consumption of pork from backyard pigs substantially. However, meat inspection had no effect on the risk after the consumption of pork from controlled housing pigs indicating that Trichinella control of controlled housing pigs has no additional value.

To assess the risk of T. solium after eating pork of pigs originating from home slaughtering in Europe, a QMRA model was developed. The model included two pathways: home slaughtering and slaughtering under meat inspection. Based on the slaughter data of five countries in Europe, it was shown that prevalence of contaminated pork portions from home slaughtered pigs was almost 14 times higher than from slaughterhouses. This was mainly influenced by the prevalence of cysticercosis in pigs. The test sensitivity of meat inspection was however low. The model also quantified that cooking can effectively reduce the risk of exposure to T. solium infected pork (Meester, Swart et al. 2019).

In conclusion, we presented three quantitative risk studies for foodborne parasites, and highlighted their utility in aiding the assessment of disease burden and potential effects of interventions. Such QMRA studies need to be carefully designed, and uncertainties present in the model must be explicitly considered. Indeed, major data gaps in the consumption phase were identified, not only frequencies of consumption or portion sizes, but also preparation habits. Also, dose-response relations remain hard to quantify. However, despite remaining challenges, QMRA studies have come a long way since their inception in the 1990s, with significant methodological advances, and have become the thefacto method of choice for microbial risk assessments.

Acknowledgements

We are grateful to the members of the Cost action Cystinet and the Cost action European network for foodborne parasites. We also thank Marina Meester for her contributions to the T. solium QMRA model.

Published in the proceedings of the International Pig Veterinary Society Congress – IPVS2020. For information on the event, past and future editions, check out https://ipvs2022.com/en.

Public health impact and control options of parasitic zoonoses via pork

What is Rendering?

E.S.E. & INTEC