First Department of Microbiology, School of Medicine, Aristotle University of Thessaloniki, Greece

ORIGINAL ARTICLE 10.1111/j.1469-0691.2005.01193.x Seroprevalence of Toxoplasma gondii in northern Greece during the last 20 years E. Diza, F. Frantzidou, E. Souliou, M. Arvanitidou, G. Gioula and A. Antoniadis First Department of Microbiology, School of Medicine, Aristotle University of Thessaloniki, Greece ABSTRACT The seroprevalence of Toxoplasma gondii in the northern Greek population was determined in 1984, 1994 and 2004, and changes during this period were investigated. In total, 1014, 812 and 958 sera from individuals aged 1 day to 70 years were examined in 1984, 1994 and 2004, respectively, for IgG and IgM anti-toxoplasma antibodies with the standard immunofluorescence assay (IFA) and microparticle enzyme immunoassay (MEIA). In individuals positive for IgM-specific antibodies, primary infection with Toxoplasma was diagnosed on the basis of the Toxoplasma serological profile (IFA, MEIA, conventional IgM and IgA ELISAs, immunosorbent agglutination assay and IgG avidity test). The prevalence of IgG-specific antibodies in the general population was 37%, 29.9% and 24.1% in 1984, 1994 and 2004, respectively, and was 35.6%, 25.6% and 20%, respectively, in women of reproductive age (15 39 years). The incidence of Toxoplasma infection, based on cases of primary infection and the annual seroconversion rate for the general population, was estimated to be 1.25% and 1.1% in 1984, 1.05% and 0.93% in 1994, and 0.85% and 0.8% in 2004. The significant decline in prevalence, and the shift towards an older age group, observed during this period could be explained by the improved socio-economic situation. The high (80%) proportion of women of reproductive age susceptible to Toxoplasma infection, with an estimated 90 200 neonates infected in utero annually, seems to present a potential risk to public health. Education of the public and prophylactic measures may become increasingly important. Keywords Immunofluorescence assay, incidence, northern Greece, prevalence, seroprevalence, Toxoplasma gondii Original Submission: 24 August 2004; Revised Submission: 27 December 2005; Accepted: 4 March 2005 Clin Microbiol Infect 2005; 11: 719 723 INTRODUCTION Toxoplasmosis is a zoonotic disease caused by Toxoplasma gondii. Although the course of infection is generally benign, this organism can cause significant morbidity and mortality in the developing foetus and in immunocompromised individuals [1]. As an effective vaccine has not yet been developed, continuous and detailed epidemiological surveillance is required to estimate the risk of infection, especially in pregnant women, and the likelihood of reactivation in immunocompromised individuals. The first seroepidemiological study of T. gondii in the regions of Macedonia and Thrace (northern Greece) was conducted in 1972, and revealed a Corresponding author and reprint requests: E. Diza, First Department of Microbiology, School of Medicine, Aristotle University of Thessaloniki, Thessaloniki 541 24, Greece E-mail: eudoxia@med.auth.gr prevalence rate of 43% [2]. The present study, performed in 2004, determined the prevalence and incidence of Toxoplasma infection in individuals aged 1 day to 70 years from the same area, as well as in the sub-population comprising women of reproductive age. The results obtained were compared with those from similar surveys performed in 1984 and 1994 [3,4]. MATERIALS AND METHODS Study population The individuals examined in 1984, 1994 and 2004 were selected randomly from the general population living in rural and urban areas of the 16 prefectures in northern Greece (population 2.8 million) [5]. The number of individuals tested from each prefecture was proportional to the population of the respective region. Serum samples were obtained from apparently healthy individuals, as follows: children who were participating in screening programmes for prevention of haemoglobulinopathies and for investigation of lipidaemic profiles; women tested before and during pregnancy; and adults who attended the hospital for an Ó 2005 Copyright by the European Society of Clinical Microbiology and Infectious Diseases

720 Clinical Microbiology and Infection, Volume 11 Number 9, September 2005 annual medical check-up or for blood donation. In total, 1014 sera were tested in 1984 (470 males, 544 females), 812 in 1994 (390 males, 422 females), and 958 in 2004 (460 males, 498 females). All participants were grouped into ten age groups: 0 0.9, 1 4, 5 9, 10 14, 15 19, 20 29, 30 39, 40 49, 50 59 and 60 70 years. Women of reproductive age (15 39 years) were also studied as a separate group. The sera were stored at ) 70 C until required for analysis. Serological methods All serum samples were examined for the presence of Toxoplasma-specific IgG and IgM antibodies separately. IgG antibodies were measured using an in-house immunofluorescence assay (IFA) [6], with results expressed in IU ml, based on the WHO first standard serum 1967 [7]. Values 12 IU ml were considered to be positive. In the case of equivocal results, the respective sera were also tested for specific IgG antibodies by microparticle enzyme immunoassay (IgG-MEIA; Abbott Laboratories, Chicago, IL, USA). IgM-specific antibodies were detected using IgM-MEIA (Abbott Laboratories). IgM-positive sera were further tested by: (1) other IgM detection methods, including an in-house IFA assay (IgM-IFA; positive = 1:50 dilution), an immunosorbent agglutination assay (biomérieux, Marcy l Etoile, France), and a Western blotting test for specific IgM antibodies against the 30-kDa peptide (MarDx Diagnostics, Carlsbad, CA, USA); (2) a sensitive capture ELISA detecting IgA-specific antibodies (Bouty SpA, Sesto S. Giovanni, Italy; positive = 10 AU ml); and (3) an assay measuring the avidity of IgG-specific antibodies (Bouty SpA), where an avidity of < 15% indicates an acute primary infection during the last 2 3 months, 15 25% indicates a primary infection during the last 6 months, and > 25% indicates an older infection. Sera were tested by both IgG-IFA and IgM-IFA, or only by IgM-MEIA, at the time of collection. The additional methods for IgM, IgA and for determining the avidity of IgG antibodies were performed in 2004 for all IgM-positive sera. Diagnostic criteria for Toxoplasma primary infection A positive IgM result has a low predictive value for identifying a primary infection with Toxoplasma [8,9]. A combination of serological assays can be used to determine the onset of infection with more certainty [10]. In the present study, a primary infection was considered to have occurred recently (i.e., during the previous 5 6 months) if there was a combination of high titres of IgG-IFA, positive IgM antibodies, and a low-avidity IgG antibody index (< 25%) in a single serum sample [8 12]. The incidence of primary infection with Toxoplasma was estimated from (1) the number of cases of primary infection year, and (2) the annual seroconversion rate of the age-specific seroprevalence [13]. Statistical methods The chi-square test was used to compare differences in prevalence rates between age groups and to assess trends over time. A p value of < 0.05 was considered to be significant. Biostatistical analysis was performed using SPSS for Windows v. 10.0.1 (SPSS Inc., Chicago, IL, USA). RESULTS Immune status against T. gondii A summary of all serological tests used and the diagnosis of immune status, based on the combined results, is shown in Table 1. The frequency of IgG-IFA anti-toxoplasma antibodies in 1984, 1994 and 2004 is shown in Table 2. In all three years, seropositivity rose gradually with age; this rise was most rapid in 1984 and became slower in 1994 and in 2004. Thus, only a small proportion of children aged 1 4 and 5 9 years (3.2% and 7.4%, respectively) were positive in 2004, and the seroprevalence then rose gradually until the age of 49 years (33.3%). A significantly higher rate (50%) was found only in the group aged 50 59 years (p 0.019), with the highest rate (56.4%) in the group aged 60 70 years. No significant difference was found between men and women, except in 2004 for the group aged Table 1. Determination of immune status for Toxoplasma gondii in the general population of northern Greece (1984, 1994 and 2004) IgG-IFA IgG avidity Year Result (IU/mL) No. positive sera IgM-IFApositive IgM-MEIApositive IgM-ISAGApositive IgM-WBpositive IgA-ELISApositive Result No. sera Immune status 1984: 1014 sera tested 1994: 812 sera tested 2004: 958 sera tested 12 800 374 9 9 9 2 3 > 25% 9 Old infection 400 3200 4 2 4 4 4 3 < 25% 4 Recent infection a < 12 634 0 ND ND ND ND Negative < 12 2 2 1 0 0 0 Negative 12 400 239 10 14 13 6 5 > 25% 14 Old infection 400 1600 3 2 3 3 3 3 < 25% 3 Recent infection a < 12 567 ND 0 ND ND ND Negative < 12 3 1 3 1 0 0 Negative 12 800 228 6 9 7 5 2 > 25% 9 Old infection 200 1600 3 3 3 3 3 3 < 25% 3 Recent infection a < 12 723 ND 0 ND ND ND Negative < 12 4 2 4 2 0 0 Negative a Primary infection acquired during the previous 6 months. IFA, immunofluorescence assay; MEIA, microparticle enzyme immunoassay; ISAGA, immunosorbent agglutination assay; WB, Western blot; ND, not determined.

Diza et al. Toxoplasma gondii in northern Greece 721 Table 2. Age-specific seroprevalence against Toxoplasma gondii in the northern Greek population Age (years) 1984 1994 2004 r n % r n % r n % 0 0.9 21 131 16 11 73 15.1 a 10 87 11.5 a 1 4 11 94 11.7 3 67 4.5 3 93 3.2 5 9 18 101 17.8 8 98 8.2 7 94 7.4 10 14 14 59 23.7 3 34 8.8 9 84 10.7 15 19 20 80 25 8 48 16.7 12 82 14.6 20 29 42 124 33.9 a 38 155 24.5 22 104 21.2 30 39 52 106 49.1 30 99 30.3 a 28 110 25.5 40 49 67 110 61 48 92 52.2 38 114 33.3 a 50 59 65 98 66.3 42 68 61.8 40 80 50 60 70 68 111 61.3 51 78 65.4 62 110 56.4 b Total number 378 1014 37.3 242 812 29.8 c 231 958 24.1 d a Significant difference compared to next age group. b Significant difference between men and women. c p 0.001 (1984 1994). d p 00059 (1994 2004). r, number seropositive; n, sample size. 60 70 years (43% in men vs. 56.4% in women; p 0.003). Overall, the seroprevalence rate in the general population in 2004 was 24.1%, which was significantly lower than the rates of 29.9% in 1994 (p 0.001) and 37% in 1984 (p 0.0059). Incidence of primary infection with Toxoplasma in the general population The two approaches used to calculate the incidence of primary Toxoplasma infection provided similar results (Table 3). Thus: (1) a primary infection was identified during the previous 6 months for four cases in 1984, three cases in 1994 and three cases in 2004, with estimated incidences of primary infection of 1.25, 1.05 and 0.83 100 seronegative individuals year in 1984, 1994 and 2004, respectively; (2) the age-specific incidence of primary infection, estimated from the annual rate of seroconversion, increased annually by c. 1.1% in 1984, 0.93% in 1994 and 0.8% in 2004. Table 3. Seroprevalence and incidence among population groups in northern Greece Population groups General population Women aged 15 39 years a Incidence year. b Incidence pregnancy. Prevalence (%) Incidence (%) Cases of primary infection Annual seroconversion rate 1984 1994 2004 1984 1994 2004 1984 1994 2004 37.3 29.8 24.1 1.25 a 1.05 a 0.85 a 1.1 a 0.93 a 0.8 a 35.6 25.6 20 0.69 b 0.67 b 0.5 b 0.8 b 0.5 b 0.4 b Prevalence and incidence of Toxoplasma infection in women of reproductive age The prevalence of Toxoplasma infection in women of reproductive age (15 39 years) was 35.6% in 1984, 25.6% in 1994 and 20% in 2004 (Table 3). Based on the number of cases of primary infection, the incidence pregnancy was estimated to be 0.69% in 1984, 0.67% in 1994 and 0.55% in 2004, while the annual seroconversion rate was 0.81% in 1984, 0.55% in 1994 and 0.48% in 2004. The mean incidence of primary infection pregnancy was 0.51% in 2004. DISCUSSION The present study showed that 24.1% of healthy individuals in northern Greece, aged 1 day to 70 years, were positive for anti-toxoplasma antibodies in 2004, and that 20% of women of reproductive age in the same population were seropositive. These seropositivity rates are significantly lower than rates reported in central European countries, but higher than those in the UK and in Scandinavian countries. Prevalence rates in the overall population have been reported to be > 50% in France [14], 52.4% in Switzerland [15], and 59% in Germany [16], while rates in women of reproductive age have been reported to be 20.3% in Finland [17], 10.9% in Norway [9], 8.1% in England [18], 40% in Switzerland [15], and 54.3% in France [19]. Differences in the prevalence of Toxoplasma infection have been associated with warm and humid environments, cooking habits and the number of cats living outdoors, but the exact roles of these factors are still not fully understood [20,21]. In northern Greece, the climate is warm and humid during the summer, but is cold during winter, and raw or undercooked meat is eaten very rarely. In 1966, Fulton et al. [22] found a high seropositivity rate of 60% in southern Greece with the Dye test, and a rate of 43% was found in northern Greece in 1972 with the same test [2]. A significant decline in the prevalence of Toxoplasma infection was observed in northern Greece in 1984 and 1994 (37.3% and 29.8%, respectively), dropping to 24.1% in 2004 in the present study. Various studies from Europe [23] and the USA [20] have also reported a decrease in the seroprevalence of Toxoplasma infection during the past 20 30 years. The reason for this is not clear, although it may

722 Clinical Microbiology and Infection, Volume 11 Number 9, September 2005 result from changes in the rate of exposure of susceptible individuals to tissue cysts or oocystcontaminated environmental reservoirs. In northern Greece, this gradual decrease may be explained by the improved socio-economic situation. Reduced consumption of home-grown fruit and vegetables, wide consumption of pasteurised milk, and the increasing use of freezers in the home, may be some of the factors that have led to a decrease in human infection, as well as increased awareness and education with respect to eating adequately cooked meat, keeping cats, hygienic vegetable preparation and hand washing [24]. No recent studies in the general population have been reported from southern Greece. In the present study, the prevalence rate in 2004 increased with age, from 3.2% in the group aged 1 4 years, to 56.4% in the group aged 60 69 years, with analogous increases in seropositivity in 1984 and 1994. There was a clear decline in the prevalence in each age group from decade to decade (Table 2). A cohort effect was observed for the group aged 20 29 years in 1984. The observed shift towards older age groups could be explained by the fact that increasingly large numbers of people in Greece live in apartments without gardens, so that children grow up away from gardens, soil and animals. However, other factors may also have influenced this shift, as different sources of infection are found in different age groups [25]. As with seroprevalence, the incidence of primary infection with Toxoplasma in the general population, as well as in women of reproductive age, seems to have decreased during the last 20 years in northern Greece. In the group of women of reproductive age, the incidence, as estimated by the cases of primary infection, decreased from 0.69% pregnancy in 1984 to 0.55% in 2004, while the incidence estimated by the annual seroconversion rate was found to have declined from 0.81% pregnancy in 1984 to 0.48% in 2004. Similar decreases have been reported from some other countries [23,26], with values ranging between 0.06% and 1.48% during pregnancy [19,20,27]. The prevalence of T. gondii infection showed a significant decreasing trend in northern Greece during the study period, resulting in a high (80%) proportion of non-immune women of reproductive age. Based on an estimated incidence in 2004 of 0.51% pregnancy, an assumed maternofetal transmission rate of 23 50% [9,20,27], and an annual birth rate of 100 000 in Greece, it can be calculated that c. 408 pregnant women are infected annually with Toxoplasma during their pregnancy, and that 90 200 children are infected in utero by T. gondii. The present results are not representative of the entire country, so further studies should be performed to establish the complete picture of Toxoplasma infection in Greece. In parallel, it is also important to inform and educate the public on how to avoid possible risk-factors and prevent Toxoplasma infection. REFERENCES 1. Hill D, Dubey JP. Toxoplasma gondii: transmission, diagnosis and prevention. Clin Microbiol Infect 2002; 8: 634 640. 2. Moraitou M. Toxoplasma and cytomegalovirus antibodies in a healthy population in northern Greece. Thesis, Aristotelian University of Thessaloniki, 1972. 3. Diza-Mataftsi E, Souliou E, Kokkini S, Frantzidou F, Dalaina V, Papapanagiotou J. Immunity in Northern Greek population to Toxoplasma gondii. Acta Microbiol Hellen 1985; 30: 85 93. 4. Diza-Mataftsi E, Frantzidou-Adamopoulou F, Souliou- Symeonidou E, Kyriazopoulou-Dalaina V, Papavasiliou P. Prenatal serological screening of toxoplasmic infection in Macedonia and Thrace. Hellen Obstet Gynecol 1994; 6: 66 73. 5. Anonymous. Statistical year book of Greece. Athens: National Statistical Service of Greece, 2002; 71. 6. Garin IP, Ambroise-Thomas P. Le diagnostic serologique de la toxoplasmose per la methode des anticorps fluorescents. Presse Med 1963; 71: 2485 2488. 7. Anonymous. WHO Expert Committee on biological standardization. WHO Tech Ser 1969; 413: 76. 8. Montoya J. Laboratory diagnosis of Toxoplasma infection and toxoplasmosis. J Infect Dis 2002; 185: S73 S82. 9. Jenum P, Stray-Pedersen B, Melby K et al. Incidence of Toxoplasma gondii infection in 35,940 pregnant women in Norway and pregnancy outcome for infected women. J Clin Microbiol 1998; 36: 2900 2906. 10. Roberts A, Hedman K, Luyasu V et al. Multicenter evaluation of strategies for serodiagnosis of primary infection with toxoplasma gondii. Eur J Clin Microbiol Infect Dis 2001; 20: 467 474. 11. Lebech M, Joynson HM, Seitz HM et al. Classification system and case definitions of Toxoplasma gondii infection in immunocompetent pregnant women and their congenitally infected offspring. Eur J Clin Microbiol Infect Dis 1996; 15: 799 805. 12. Lappalainen M, Koskela P, Koskiniemi M et al. Toxoplasmosis acquired during pregnancy: improved serodiagnosis based on avidity of IgG. J Infect Dis 1993; 167: 691 697. 13. Punda-Polic V, Tonkic M, Capkun V. Prevalence of antibodies to Toxoplasma gondii in the female population of the county of Split Dalmatia, Croatia. Eur J Epidemiol 2000; 16: 875 877. 14. Ambroise-Thomas P, Schweitser M, Pinon JM, Thiebaugeorges O. Prevention of congenital toxoplasmosis

Diza et al. Toxoplasma gondii in northern Greece 723 in France. Risk assessment. Results and perspectives of prenatal screening and newborn follow up. Bull Acad Natl Med 2001; 185: 665 683. 15. Jacquier P, Nadal D, Zuber P, Eckert J. The status of infection with Toxoplasma gondii in the Swiss population: contribution of a seroepidemiologic study from the Zurich canton. Schweiz Med Wochenschr Suppl 1995; 65: 23S 28S. 16. Fiedler K, Hulsse C, Straube W, Briese V. Toxoplasmosisantibody seroprevalence in Mecklenburg-Western Pomerania. Zentralbl Gynakol 1999; 121: 239 243. 17. Lappalainen M, Koskela P, Hedman K et al. Incidence of primary toxoplasma infections during pregnancy in southern Finland: a prospective cohort study. Scand J Infect Dis 1992; 24: 97 104. 18. Allain JP, Palmer CR, Pearson G. Epidemiological study of latent and recent infection by Toxoplasma gondii in pregnant women from a regional population in the UK. J Infect 1998; 36: 189 196. 19. Ancelle T, Goulet V, Titard-Fleury V et al. La toxoplasmose chez la femme enceinte en France en 1995. Bull Epidemiol Hebdomaire 1996; 51: 227 228. 20. Remington JS, McLeod R, Thulliez P, Desmonts G. Toxoplasmosis. In: Remington JS, Klein JO, eds. Infectious diseases of the fetus and newborn infant, 5th edn. Philadelphia: WB Saunders, 2001; 205 346. 21. Lynfield R, Guerina NG. Toxoplasmosis. Pediatr Rev 1997; 18: 75 83. 22. Fulton JD, Fleck DG, Payhe RA. Prevalence of toxoplasma antibodies in sera from Greece and Africa. J Hyg (Lond) 1966; 64: 75 79. 23. Ho-Yen DO. Clinical features. In: Ho-Yen DO, Joss AWL, eds. Human toxoplasmosis. Oxford: Oxford Medical Publications, 1992; 56 78. 24. Cook AJ, Gilbert RE, Buffolano W et al. Sources of toxoplasma infection in pregnant women: European multicentre case-control study. European Research Network on Congenital Toxoplasmosis. BMJ 2000; 321: 127 128. 25. Baril L, Ancelle T, Goulet V, Thulliez P, Tirard-Freury V, Carme B. Risk factors for Toxoplasma infection in pregnancy: a case-control study in France. Scand J Infect Dis 1999; 31: 305 309. 26. Foulon W, Naessens A, Derde MP. Evaluation of the possibilities for preventing congenital toxoplasmosis. Am J Perinatol 1994; 11: 57 62. 27. Couvreur J. Problems of congenital toxoplasmosis. Evolution over four decades. Presse Med 1999; 28: 753 757.