Toxoplasma gondii Infection in South-East Europe : Epidemiology and Epizootiology

Toxoplasmosis is a globally distributed parasitic zoonosis. Cats and other Felidae are the definitive host of the Toxoplasma gondii parasite, in whose intestines the sexual cycle takes place, and they are the primary source of infection to all animals including humans, by way of contaminating the environment with oocysts excreted in the feces (Dubey, 2010). Herbivores are infected by ingestion of food and water contaminated by oocysts, and carnivores by eating tissue cysts present in the flesh of infected animals. Omnivores including humans are infected by both routes – by oocysts via improperly washed vegetables or fruits, contaminated water or hands, and by tissue cysts via improperly processed or raw meat. Vertical transmission, from mother to offspring, may occur, and is the cause of congenital toxoplasmosis (CT). CT is actually the major T. gondii-induced clinical entity, which, along with opportunistic infection in immunocompromised patients, defines the clinical significance of toxoplasmosis. The gravity of the potential consequences of CT on the one hand, and the preventability of the disease on the other, call for implementation of prevention programs. A prerequisite for an adequate choice of prevention strategy is continuous monitoring of the local epidemiological situation.


Introduction
Toxoplasmosis is a globally distributed parasitic zoonosis. Cats and other Felidae are the definitive host of the Toxoplasma gondii parasite, in whose intestines the sexual cycle takes place, and they are the primary source of infection to all animals including humans, by way of contaminating the environment with oocysts excreted in the feces (Dubey, 2010). Herbivores are infected by ingestion of food and water contaminated by oocysts, and carnivores by eating tissue cysts present in the flesh of infected animals. Omnivores including humans are infected by both routes -by oocysts via improperly washed vegetables or fruits, contaminated water or hands, and by tissue cysts via improperly processed or raw meat. Vertical transmission, from mother to offspring, may occur, and is the cause of congenital toxoplasmosis (CT). CT is actually the major T. gondii-induced clinical entity, which, along with opportunistic infection in immunocompromised patients, defines the clinical significance of toxoplasmosis. The gravity of the potential consequences of CT on the one hand, and the preventability of the disease on the other, call for implementation of prevention programs. A prerequisite for an adequate choice of prevention strategy is continuous monitoring of the local epidemiological situation.
This chapter reviews the epidemiology and epizootiology of T. gondii infection in South-East Europe (SEE). SEE is here considered as the territory comprising the Balkan Peninsula (35° -46°53' N latitude, 13°23' -30° E longitude), bordered by the Adriatic Sea to the west, the Mediterranean Sea to the south, the Black Sea to the east and the rivers Sava and Danube to the north, encompassing the countries descending from ex-Yugoslavia including Slovenia, Croatia, Bosnia & Herzegovina, Serbia, Montenegro and FYR of Macedonia (FYRoM), as well as Albania, Bulgaria and mainland Greece. Most of the area is mountainous, while the climate varies from Mediterranean to moderate. The whole region has a combined area of 550,000 km 2 and a population of 55 million.

Toxoplasmosis in generative age women
Data on the prevalence of T. gondii infection in SEE countries are presented in Figure 1. In the last ten years, the prevalence has not surpassed 50% anywhere in the region, ranging from 20% in Greece (Diza et al., 2005) to 49% in Albania (Maggi et al., 2009). Wide differences in the prevalence of infection are generally characteristic of Europe, since the infection prevalence is currently ranging from 8.2% in Switzerland (Lausanne and Geneva) (Zufferey et al., 2007) to 57.6% in Timisoara-Romania (Olariu et al., 2008). Differences in the prevalence of T. gondii infection are commonly explained by differences in life-style habits pertaining to risk factors for transmission in particular milieus. Not many studies on infection risk factors have been published in SEE countries, but those available identified consumption of undercooked meat as the leading risk factor for transmission in Serbia and, more recently, in Albania (Bobić et al., 1998;Maggi et al., 2009), and contact with soil in northern Greece and FYRoM (Decavalas et al., 1990;Diza et al., 2005;Cvetković et al., 2010). Exposure to soil was also considered to account for the higher prevalence of infection in rural vs. urban women in Croatia.
Continuous monitoring of the prevalence of T. gondii infection in women of childbearing age in Slovenia, Serbia and Greece has showed a significant decrease in the infection prevalence since the eighties onwards. The largest decrease, from 86% in 1988 to 31% in 2007 (Bobić et al., 2003;, was noted in Serbia. Furthermore, a trend of decreasing prevalence has been shown during the last decade in FYRoM, from 25% in 2002 to 20% in 2005 (Cvetković et al., 2003;2010) and in Montenegro, a rather dramatic one, from 41% in 2001 to 27% in 2007 (Mišković et al., 2003;Rajković & Vratnica, 2008). These data suggest that a decrease in the prevalence is a region-wide feature (Fig. 1    to such a change, including increased public awareness as a result of health education, better hygiene on livestock farms, and more frequent use of frozen meat. However, according to the SEE data, the listed factors seem not to be exhaustive; although better farming conditions along with the increased consumption of frozen meat (freezers now present in most households) may have contributed to a decrease in the infection prevalence in Serbia and Albania, a decreasing trend was also noted in FYRoM and Greece where consumption of undercooked meat was not found to be a risk factor.
It appears that, according to reports from the eastern part of the region, there is a north-tosouth decrease in the infection prevalence (Table 1). For instance, in 1994, the prevalence ranged from 69% in southern Hungary (as a region neighbouring the SEE to the north) (Szénási et al., 1997), over 53% in Serbia (Bobić et al., 2003) to 26% in northern Greece (Diza et al., 2005). This trend was also evident within SEE in all years for which comparative data were available ( Seasonality of infection was examined in Slovenia (Logar et al., 2005) in the west and Serbia  in the east. Both studies showed a strong seasonality, with significantly more cases of acute infection in the winter than in the summer months (Fig. 2). In Slovenia, seasonality of infection was attributed to "more frequent and closer contacts with potentially T. gondii infected cats, which prefer to stay indoors during this period" (Logar et al., 2005). In contrast, more cases of acute infection in the winter in Serbia were explained by a higher influence of undercooked meat consumption in the winter period . Accordingly, the decrease in the prevalence of T. gondii infection in the SEE seems independent of the varying influence of risk factors for infection transmission throughout the region. However, this may prove not to be entirely true, as the above analysis was based on the limited data reported, which were often not acquired through systematic nation-wide research, and which were mainly derived from epidemiological questionnaires of variable precision level. For an accurate insight into the risk factors of major significance for human T. gondii infection, more research, based on larger patient series and carried out in different SEE areas, including case-control studies, would be preferable.

Toxoplasmosis in immunocompromised patients
Toxoplasmosis is a major opportunistic infection causing life-threatening disease in immunocompromised individuals, which is considered to be a consequence of reactivation of previously latent infection. Although opportunistic infection due to T. gondii has long been known in organ/tissue transplantations and patients with malign or systemic diseases on treatment with immunosuppressive effect, data on toxoplasmosis in the immunocompromised population in SEE are available only from groups of HIV-infected individuals and even these are scarce.

Molecular epidemiology
Although the population structure of T. gondii isolates throughout the world is currently a major research interest in the field of toxoplasmosis ( . It is to be hoped that with the increased use of molecular methodologies in the region, work will be performed which will contribute to the pan-European map of T. gondii genotypes.

Prevention of human toxoplasmosis
Strategies for the prevention of CT include general screening-in-pregnancy programs and health education, and countries with a low prevalence of infection generally opt for health education (Kravetz et Federman, 2005;Gilbert and Peckham, 2002), while those with a high prevalence adopt screening-in-pregnancy programs (Aspöck & Pollak, 1992;Thulliez, 1992).
In the SEE, a systematic program for the prevention of CT based on serological screening of pregnant women and health education has been implemented in Slovenia, while no other country has a systematic prevention program (Logar et al, 2002). The changing pattern of infection across the region currently complicates the choice of prevention strategy. A decrease in the prevalence of infection in women of childbearing age implies a rising proportion of women susceptible to infection in pregnancy, which, in turn, may lead to an increase in the incidence of congenital infections. Indeed, an increase in the incidence of primary infections in pregnancy has been shown through systematic screening-in-pregnancy programs; in Austria, which has been screening all pregnant women ever since 1975, the decrease in T.
gondii prevalence from 50% to 37% during the 1980s was followed by an increase in the incidence of primary infections in pregnancy from 0.4% to 0.8% (Aspöck & Pollak, 1992). Similarly, in Slovenia, the decrease in the prevalence of infection from the early 1980s onwards, therefore before the introduction of the systematic prevention program in 1995, has been associated with an increased incidence of infections in pregnancy from 0.33% in the early 1980s to 0.75% in the early 1990s and to 0.94% in the late 1990s (Logar et al., 1995;. The decreasing prevalence of T. gondii infection and a possible subsequent increase in the incidence of primary infections in pregnancy warrants introduction of CT prevention programs in the SEE countries. A prerequisite for such programs to be cost-effective, however, is an accurate assessment of the proportion of women of childbearing age susceptible to infection in pregnancy and the subsequent incidence of CT. These data are not yet available for most of the SEE, and are further complicated by the changing pattern of infection across the region. Thus, a sound and financially sustainable alternative that may be recommended for all of the SEE, similar to what has been recommended for Serbia (Bobić et al., 2003), includes health education of all women of childbearing age focusing on the locally significant risk factors for infection transmission in particular countries.
Health education is an adequate preventive measure for uninfected immunocompromised patients too.

Epizootiology of toxoplasmosis in SEE
Epizootiological surveillance of T. gondii infection in meat animals is important both as a measure essential for animal health and economics of livestock production, and for assessment of the risk for human health, as infected meat animals represent an important reservoir for human infection.
Throughout SEE, few data are available on T. gondii infection in animals and most are not the result of systematic research. Early reports, in particular from Serbia/Yu, dealt with isolation of viable T. gondii from various species. By bioassay of brains and/or hearts in ground squirrels (Spermophilus citellus) T. gondii was isolated from 0.25% chickens (Simitch et al., 1961); and 0.7% pigs, 0% sheep and cattle, 6.2% guinea fowl, 3.8% turkeys, 3.6% ducks, 1% geese and 0% pigeons (Simić et al., 1967). By mouse bioassay, T. gondii has been isolated from pork, and from pig and sheep diaphragms in Croatia/Yu (Wikerhauser et al., 1983;. In Bosnia/Yu, Živković and Arežina (1991) reported apparent T. gondii in smears from muscle tissues of farmed layers.
The data on strain isolation in animals are scarce; researchers in Bulgaria have reported isolation of both virulent and avirulent strains of T. gondii from domestic and wild animals, but did not perform genetic typing (Arnaudov, 1978;Arnaudov et al., 2003;Arnaudov & Arnaudov, 2005). In Serbia, genotyping of T. gondii strains isolated from animals (sheep, pigs, pigeons) is in progress, and the first results confirm the dominance of type II strains (unpublished data).
Earlier serological investigations of various animal species in SEE countries showed a prevalence of 37% for cattle, 30% for sheep, 26% for pigs, 17% for horses, 41% for dogs, 25% for cats, 52% for house mice and 20% for rats in Serbia, but except for pigs, they were carried out on samples of limited size (Šibalić, 1977 (Table 3), possibly associated with a comparably increased humidity in this region . The levels of specific antibody determined in the cattle were relatively low (not above 1:400), which is consistent with reports on the rapid decrease of T. gondii antibody in cattle (Dubey et al., 1985). Cattle are relatively resistant to T. gondii infection (Dubey and Thulliez, 1993), but it is unclear whether this is associated with fast elimination of cysts from cattle tissues (Dubey, 1983), or there is inconsistent cyst formation following infection. In addition to farm location in the western region, the only other risk factor determined for cattle infection was small herd size. According to the type of housing, the results showed that access to outside pens was protective as compared to total confinement. This apparently paradoxical finding may represent a further argument for the importance of the way in which feed is kept, or possibly indicates involvement of other farm factors not identified in this type of study.
The same study showed a prevalence of 85% in sheep in Serbia, of which 10% had high antibody levels of ≥1:1600, suggestive of acute infection. Although arbitrary, the cut-off of 1:1600 is even conservative since Dubey and Welcome (1988) had considered a titre of 1:1024 high. However, correlation with ovine abortions could not be established, since etiological laboratory diagnosis of ovine abortions in Serbia does not include diagnosis of T. gondii. Regionwise, similarly as with cattle, sheep from Western Serbia were at an increased risk of infection as compared to all other regions (Table 3). An increased risk of infection was also found in state-owned vs. private large flocks. Compared to other SEE countries, the prevalence of 85% is quite high; in Croatia and Bulgaria the highest recorded prevalence in sheep is 48%, and in Greece 58.5% (Table 2). This is also evident in goats; compared with the prevalence of 74. In pigs, an overall seroprevalence of 29% was established in Serbia . Of those seropositive, 4% were likely to be in the acute stage of infection, indicating continuous presence of infection reservoirs in the environment. Risk factors included age and farm type (Table 3). Since pigs are continuously exposed to infection, the increase in the risk of infection with age, ranging from 15% in market weight age pigs to 41% in adults, was expected, and repeated previous findings (Dubey et al., 1991;Dubey et al., 1995;Weigel et al., 1995;Damriyasa et al., 2004). Pigs on finishing type farms were four-fold more likely to be infected than those from farrow-to-finish farms. According to these results, it was proposed that a national strategy to reduce the level of T. gondii infection in pigs should include a shift towards the development of more farrow-to-finish farms, as well as vigilance in farm management and implementation of zoo-hygienic measures at finishing farms. Damriyasa et al. (2004) stated that T. gondii seropositivity is an indicator of the hygienic status of the pig farm. vs. those from farrow-to-finish intensive farms.
On the other hand, the modern approach in farm management to provide for the welfare of the animals as well as organic food for human consumption is to develop animal-friendly (organic) farms. According to experiences from the Netherlands (Kijlstra et al., 2004), development of such farms may result in an increase in T. gondii infection. Nevertheless, a single report from organic sheep and goat farms in Greece (Ntafis et al., 2007) showed similar prevalence rates to those in animals from conventionally managed farms (Kontos et al., 2001;Diakou et al., 2005b).
A major reason for the control of T. gondii infection in meat animals is the reduction of the reservoir of human infection. Cattle are generally thought not to be significant in this context (Dubey and Thulliez, 1993). However, beef is often consumed undercooked ('rare' beef steaks, roast beef, steak tartar), and at least one outbreak of toxoplasmosis whose source was raw beef has been documented (Smith, 1993 For most meat animals, although a trend is generally (worldwide) difficult to establish due to the scarcity of studies in most countries (no two time points), there is no visible reduction in the prevalence of T. gondii infection, as opposed to the decreasing trend in humans, discussed in detail earlier in this chapter (and explained by reasons including increased frozen meat use, better farm management etc.). For the most part, farming practices and environmental contamination have not changed, and except for the intensive pig farms in which a major reduction in T. gondii prevalence has occurred, a decline in the prevalence of T. gondii infection in meat animals is yet to be achieved. Moreover, for strictly herbivorous species that require outdoor access, this is probably impossible (Kijlstra & Jongert, 2009).

Conclusion
Existing strategies for the prevention of toxoplasmosis in countries which have been implementing them for years have led to a decrease in its incidence, but have not solved the problem of congenital infection. This clearly shows that new comprehensive strategies for the prevention of toxoplasmosis are needed. These should be based on accurate and validated data on (1) the routes and risk factors for human infection on local level, which will allow for a more efficient health education; (2) routes and risk factors for meat animal infection to diminish infection reservoirs; (3) environmental contamination. Epidemiological and epizootiological data presented in this chapter show how far along this road we have come, and more importantly, how far we still have to go to achieve successful prevention of