Toxoplasmosis among pregnant women in France: Risk factors and change of prevalence between 1995 and 2003

Revue d’E´pide´miologie et de Sante´ Publique 57 (2009) 241–248

Original article

Toxoplasmosis among pregnant women in France: Risk factors and change of prevalence between 1995 and 2003 La toxoplasmose chez la femme enceinte en France : facteurs de risque associés et évolution de la prévalence entre 1995 et 2003 F. Berger a,b,*, V. Goulet a, Y. Le Strat a, J.-C. Desenclos a b

a De´partement maladies infectieuses, Institut de veille sanitaire, Saint-Maurice, France Institut Pasteur de la Guyane, 23, avenue Pasteur, BP 6010, 97306 Cayenne cedex, France

Received 31 October 2008; accepted 25 March 2009 Available online 3 July 2009

Abstract Background. – Congenital toxoplasmosis may affect any organ, produce severe complications such as hydrocephalus and lead to ocular lesions which can appear late after birth. Fetal outcome can be fatal. During pregnancy, the mother can become infected, particularly by eating undercooked meat or poorly washed raw fruits and vegetables. In France, prevention of congenital toxoplasmosis is based on serial serology tests, performed repeatedly until delivery for pregnant women who remain negative then at the first prenatal visit. Methods. – We used toxoplasmosis serology data collected during national perinatal surveys conducted in 1995 and 2003 to estimate toxoplasmosis prevalence, assess risk factors and ascertain time trends in prevalence. In each survey, socioeconomic variables as well as the serological status for toxoplamosis were collected for all pregnant women who had terminated a pregnancy during a given week period. Results. – In 2003, 15,108 pregnant women were included; the prevalence of Toxoplasma infection was 43.8% (95% CI: 43.0–44.6). Prevalence increased with age and was greater for those who lived in the Southwest of France, the greater Paris area and in overseas districts. In 2003, as in 1995, prevalence increased with education level, occupational status of the household and number of pregnancies. Prevalence decreased 19% between 1995 and 2003. The decrease was significantly greater for pregnant women aged less than 30 years than for those aged more or equal to 30 years. Conclusion. – Although it decreased over time, toxoplasmosis prevalence remained higher in France than in other European countries. With more than one out of two women in France susceptible of having Toxoplasma infection, it is important to promote preventive measures to avoid infection during pregnancy. # 2009 Elsevier Masson SAS. All rights reserved. Keywords: Toxoplasmosis; Prevalence; Pregnant women; Associated factors; Trends over time

Re´sume´ Position du proble`me. – La toxoplasmose conge´nitale (TC) peut atteindre tous les organes, entraıˆner des atteintes neurologiques graves (hydroce´phalie, microce´phalie) et des atteintes oculaires pouvant n’eˆtre diagnostique´es que tardivement apre`s la naissance. La TC est potentiellement mortelle pour le fœtus. Lors de la grossesse, la me`re peut s’infecter par ingestion de viande parasite´e insuffisamment cuite ou de crudite´s mal lave´es. La pre´vention de la toxoplasmose conge´nitale en France impose aux femmes enceintes se´rone´gatives une surveillance se´rologique depuis la de´claration de grossesse jusqu’a` l’accouchement. Me´thodes. – Les donne´es de se´rologie toxoplasmose de l’ensemble des femmes ayant termine´ une grossesse en France ont e´te´ recueillies a` l’occasion d’enqueˆtes pe´rinatales re´alise´es durant une semaine en 1995 et en 2003. Les donne´es de chacune des deux e´tudes ont permis d’estimer la pre´valence de la toxoplasmose et son e´volution dans le temps. Des donne´es socioe´conomiques ont permis d’identifier des facteurs associe´s a` la maladie. Re´sultats. – En 2003, 15 108 femmes enceintes ont e´te´ inclues ; la pre´valence de la toxoplasmose e´tait de 43,8 % [43,0–44,6]. La pre´valence augmentait avec l’aˆge et de´pendait de la zone ge´ographique. Elle e´tait e´leve´e dans le Sud-Ouest, en re´gion parisienne et dans les de´partements

* Corresponding author. E-mail address: [email protected] (F. Berger). 0398-7620/$ – see front matter # 2009 Elsevier Masson SAS. All rights reserved. doi:10.1016/j.respe.2009.03.006

242

F. Berger et al. / Revue d’E´pide´miologie et de Sante´ Publique 57 (2009) 241–248

d’outre-mer. En 2003, comme en 1995, la pre´valence augmentait avec le niveau d’e´tude, la situation professionnelle du me´nage et le nombre de grossesses. La pre´valence a chute´ de 19 % entre 1995 et 2003. Cette de´croissance est plus importante chez les femmes de moins de 30 ans que chez celles plus aˆge´es. Conclusion. – Malgre´ une baisse de la pre´valence de la toxoplasmose, cette dernie`re reste une des plus e´leve´e d’Europe. Il faut cependant noter que plus d’une femme sur deux est susceptible d’eˆtre contamine´e lors de sa grossesse. Pour cette raison, il est important de bien promouvoir les mesures de pre´vention a` respecter pendant la grossesse afin d’e´viter une infection durant cette pe´riode. # 2009 Elsevier Masson SAS. Tous droits re´serve´s. Mots cle´s : Toxoplasmose ; Pre´valence ; Femme enceinte ; Facteurs associe´s ; E´volutions

1. Introduction Toxoplasma gondii, the causal agent of toxoplasmosis, is an intra-cellular protozoan that has three infectious stages: tachyzoites, bradyzoites and sporozoites [1]. The parasitic cycle includes a sexual cycle in the final host (cats and other felines) and an asexual cycle in the intermediate host (homeotherms). Cats excrete oocysts in their feces. After 1–5 days, the oocysts contain sporozoites and become infectious (sporulation). Humans can be contaminated by eating foods contaminated by oocysts, such as vegetables or fruits and also by ingesting contaminated water [2,3]. Contamination can also occur by eating undercooked meat containing cysts. Young children and adults with poor hygiene after gardening or soil manipulation can be contaminated by accidentally ingesting contaminated dirt. Among pregnant women, primary infection can lead to congenital toxoplasmosis (CT). CT may cause a wide variety of manifestations and is potentially fatal for the fetus. Most children with CT are asymptomatic at birth, but cases with serious neurological complication such as hydrocephalus can be observed. About 25% will develop ocular manifestations (retinochorioditis) that may be diagnosed late. Since 1978, the French health authorities have implemented a prevention program for CT which recommends systematic serologic testing of non-immune women under 50 before marriage; since 1992 the recommendations include compulsory monthly serological screening of seronegative pregnant women, from diagnosis of pregnancy until delivery. In the event of a seroconversion in a seronegative pregnant mother, an antenatal diagnosis is proposed unless seroconversion is diagnosed in late pregnancy. The antenatal diagnostic procedure consists of fetal ultrasound for detecting fetal abnormalities, amniocentesis for PCR to search for parasitic DNA and, mouse inoculation, which is the standard test used to diagnose fetal contamination. A neonatal diagnostic procedure (serology) is also carried out at birth on blood samples drawn from the mother, the child and the umbilical cord. A placenta sample is used for mouse inoculation to establish the neonatal diagnosis. In the event of seroconversion during pregnancy, treatment with spiramycin is proposed to the mother until delivery. If fetal contamination is demonstrated, spiramycin, which is ineffective in fetal lesions, is replaced by pyrimethamine combined with sulfadiazine or sulfadoxine. For the year 1995, the annual number of new cases of CT has been estimated between 400 and 800, of which 100 to 200 may

develop long-term complications, mainly ocular manifestations [4]. Toxoplasmosis seroprevalence among pregnant women was about 80% in the 1960s [5] and about 66% [6] in the 1980s. Seroprevalence varies by geographic region, with prevalence below 30% observed mainly in North America [7], Great Britain [8], Scandinavia [9] and Southeast Asia [10]. Seroprevalence above 60% is found mainly in Africa [11– 14] and in Latin America [15,16]. In France, there is a strong regional disparity. A regional difference was also observed in the USA and in the Netherlands [7,17]. Geoclimatic factors (temperature, hygrometry, altitude) may be the source of such differences, as well as different eating behaviors [7]. In France, a positive correlation between prevalence of past infection in humans and geographic areas where sheep is the predominant source of meat consumed has been observed [18]. In the present study, we:  estimated the prevalence of Toxoplama infection in 2003 among women of childbearing age;  analyzed factors associated with Toxoplasma infection;  compared the prevalence between 2003 and 1995 in order to assess changes in prevalence over time. 2. Subjects and methods 2.1. National perinatal surveys Toxoplasmosis data were incorporated into the National Perinatal Survey conducted in 2003 (ENP 2003) and in 1995 (ENP 1995). The 2003 survey, the third carried out under the French National Perinatal Plan, aimed at:  monitoring the main health indicators for pregnant women;  evaluating medical practices during pregnancy and delivery;  identifying perinatal risk factors. Based on cross sectional survey of French births, the survey covers the entire population of women who terminated a pregnancy during a given week in France. Data are collected using a standardized questionnaire, which is administered individually to each delivering mother included. 2.2. Population For each of the three perinatal surveys, the study population consisted of all the women in late pregnancy admitted to

F. Berger et al. / Revue d’E´pide´miologie et de Sante´ Publique 57 (2009) 241–248

obstetrics, gynecology (late abortions) and surgery (cesarean sections and late abortions) units throughout Metropolitan France and overseas districts between midnight, Monday, October 13, 2003 and midnight, Sunday, October 19, 2003. The mothers of children born outside these hospital units (at home or elsewhere) and then transferred to an obstetric ward were also included. Women who delivered live or stillborn children after at least 22 weeks of amenorrhea and women whose child or fetus weighed at least 500 g at birth were also included. A standardised questionnaire was completed by the ward medical staff for each birth. In the event of multiple births, only a single questionnaire was included in the analysis. Sociodemographic characteristics of the mothers and data on prenatal care were obtained by interviewing mothers before they left the maternity ward. Mothers who could not read or understand French were interviewed via a friend or a relative. In the event of a refusal to collect the information by the ward medical staff or a refusal of mothers to participate, the information collected on the 8th day postpartum health certificate were used to gather information on sociodemographic variables included in the questionnaire and was completed by further information from medical records. A woman was considered seronegative if no anti-Toxoplasma antibodies at the last available test during pregnancy were indicated in her medical records. She was considered seropositive if the last serological test indicated the presence of specific antibodies (IgG and IgM assay). The following variables were available for analysis in the 1995 and 2003 survey: age, number of pregnancies (including the current pregnancy), education level, geographic area of residence, nationality (France, other country in Europe, North Africa, other country in Africa and any other country), marital status, occupational status of the woman or the couple (the highest occupational category of the two parents when the woman was married or lived with a partner) and Toxoplasma serology. 2.3. Statistical analysis We used the prevalence ratio and its 95% confidence interval (95% CI) to compare prevalence rates by mothers’ characteristics. The Chi2, Student’s t-test and the Kruskal–Wallis test were used for statistical testing as appropriate. Seroprevalence rates were compared by region of residence after direct age standardization using the national 2003 population as the reference. We then used multiple logistic regression to assess the independent association of Toxoplasma infection with variables which were associated with Toxoplasma infection in the univariate analysis with a p-value < 0.01. The choice of such a low p-value for variable inclusion was related to the large sample size. A similar criterion had been used for previous analyses of this survey for other outcomes [19,20]. STATATM9.0 was used for the statistical analysis. Significant interactions existed between nationality and marital status, occupational status, place of residence and age. For this reason, the multivariate analysis was stratified by

243

nationality including only women of French nationality and women from North Africa (the number of women of other nationalities were too small). The logistic model was constructed through a descending step-by-step procedure using data restricted to women of French nationality. The variables ‘‘couple’s occupational status’’ and ‘‘education level’’ were correlated, and excluding one variable of the model resulted in a significant association between the other one and Toxoplasma infection. No interaction was found between these two variables in French women ( p = 0.3). Based on these findings and taking into consideration studies conducted in the Netherlands and the USA in which an association between toxoplasmosis infection and education and socioeconomic levels had been observed, we forced the two variables into the final model. Since odds-ratio (OR) may overestimate prevalence ratios when the outcome is greater than 5 to 10%, we used a modified Poisson model to estimate them. This model with robust error variance is a recommended alternative to estimate prevalence ratios [21,22]. The modified Poisson model was applied to variables independently associated with Toxoplasma infection and identified by the logistic regression model. In order to assess changes in prevalence between 1995 and 2003, we compared the prevalence of Toxoplasma infection in 2003 to the one observed in 1995 by multiple logistic regression to adjust for sociodemographic changes over time, since similar variables had been collected in 1995 and 2003. This comparison was restricted to French mothers. 3. Results In 2003, the sample consisted of 15,378 children and 15,108 mothers. The average age of the mothers was 29.2 years [14–54]. It varied by region with an average of 28.3 years in the North and 30.2 years in the greater Paris area. For 33.9% of the women, this was their first pregnancy (Table 1). The proportion of women of French nationality was 87.9%. Almost 44% of delivering women had at some time in their lives been infected with Toxoplasma. The prevalence increased with age from 31.0% for women aged 14 to 19 to 58.2% for those aged 40 to 54 (Table 1, Chi2 test for linear trend, p < 0.001). It also increased with the number of pregnancies and education level (Table 1). The prevalence varied substantially by French region being lowest in Eastern France (29.5%) and greatest in overseas departments (54.8%) and in the greater Paris area (52.7%) (Fig. 1). It differed also significantly by mother’s nationality: 44.0% for women of French nationality, 48.5% for women from Sub-Saharan Africa, 48.3% for those from North Africa and 34.4% for those of countries outside Europe and outside Africa (Table 1). Seroprevalence was significantly greater among women with the highest occupational category (50.6%, Table 1). 3.1. Multivariate analysis We found a significant interaction between nationality and age ( p < 0.001) with prevalence increasing with age among

F. Berger et al. / Revue d’E´pide´miologie et de Sante´ Publique 57 (2009) 241–248

244

Table 1 Toxoplasmosis seroprevalence by demographic characteristic in a population of pregnant women in the French National Perinatal Survey, France, 2003. 2003

N (15,108)

Seroprevalence N

%a

Prevalence ratio

Serology Age (in years) < 20 20–24 25–29 30–34 35–39 > 39

14,704

6,443

43.8

435 2,411 4,899 4,709 1,979 415

420 2,333 4,793 4,600 1,931 399

31.0 33.0 40.7 47.2 56.9 58.2

1.0 1.1 1.3 1.5 1.8 1.9

Total

14,848

14,476

5,098 9,928

4,967 9,683

39.4 46.1

1.0 1.2

15,026

14,650

5,960 3,108 4,698 565

5,869 3,049 4,595 537

46.8 41.5 42.4 39.7

1.0 0.9 0.9 0.8

14,331

14,050

1,134 1,667 1,833 1,036 2,391 1,582 1,320 3,197 609

1,117 1,602 1,816 1,025 2,330 1,503 1,295 3,113 595

29.5 36.0 37.2 41.2 42.9 45.8 50.0 52.7 54.8

1.0 1.2 1.3 1.4 1.5 1.5 1.7 1.8 1.9

Total

14,769

14,396

Nationality France Other Europe North Africa Other Africa Other

12,853 377 611 354 421

12,622 364 580 338 395

44.0 35.7 48.3 48.5 34.4

1.0 0.8 1.1 1.1 0.8

Total

14,616

14,299

1,247 5,405 7,825

1,199 5,310 7,662

42.5 41.8 45.6

1.0 1.0 1.1

14,477

14,171

2,561 1,329 2,716 5,309 1,745 806

2,516 1,298 2,678 5,198 1,700 768

50.6 44.7 44.8 41.5 41.2 42.1

1.0 0.9 0.9 0.8 0.8 0.8

14,466

14,158

1.1–1.2

< 0.001

Education level > Baccalaureate Last year of high school Secondary school Elementary school Total Economic zone (ZEAT) b East Center-East West North Paris Basin Mediterranean Southwest Greater Paris Overseas departments

0.8–0.9 0.8–0.9 0.8–0.9

< 0.001 1.1–1.4 1.1–1.4 1.2–1.6 1.3–1.6 1.4–1.7 1.5–1.9 1.6–2.0 1.7–2.1

< 0.001

Marital status Single Unmarried couples Married couples or couples in Civil Solidarity Pacts (PACS) Total

a

0.9–1.2 1.1–1.5 1.3–1.8 1.6–2.1 1.6–2.2

< 0.001

Total

Total

p < 0.001

Number of pregnancies One At least two

Occupational status of couple ‘‘Executives’’ Trade services/merchants Intermediate occupations Employees Manual workers No occupation

95% CI

0.7–0.9 1.0–1.2 1.0–1.2 0.7–0.9

< 0.001 0.9–1.1 1.0–1.2

c

< 0.001 0.8–0.9 0.8–0.9 0.8–0.9 0.8–0.9 0.8–0.9

Percentage of women testing positive. East (Alsace, Franche-Comte´, Lorraine), Center-East (Auvergne, Rhoˆne-Alpes), West (Bretagne, Pays-de-la-Loire, Poitou-Charentes), North (Nord-Pas-deCalais), Paris Basin (Bourgogne, Centre, Champagne-Ardenne, Basse- and Haute-Normandie, Picardie), Mediterranean (Languedoc-Roussillon, Provence-AlpesCoˆte d’Azur, Corse), Southwest (Aquitaine, Limousin, Midi-Pyre´ne´es), Greater Paris (Iˆle-de-France), overseas departments (Martinique, Guadeloupe and Guyane). c ‘‘Highest’’ occupation of the two partners (current occupation or last occupation held). b

F. Berger et al. / Revue d’E´pide´miologie et de Sante´ Publique 57 (2009) 241–248

245

30 years or over (results obtained using the modified Poisson regression model yielded similar estimates, data not shown). 4. Discussion

Fig. 1. Toxoplasmosis seroprevalence standardized by age among pregnant women, according to eight regions (ZEAT, Insee definition), in the French National Perinatal Survey 2003.

French mothers and those from North Africa. However, this was not observed for mothers of other origin. For women of French nationality, the region of residence ( p < 0.001), the age group ( p < 0.001) and number of pregnancies ( p = 0.04) were independently associated with Toxoplasma infection (Table 2). For women of North African origin, none of the variables were associated with Toxoplasma infection (data not shown). 3.2. Comparison of Toxoplama infection between 1995 and 2003 In 2003, the women included in the survey were on average older than those in 1995 (29.2 years versus 29.0 years of age, p < 0.001) and seroprevalence was lower (54.3% versus 43.8%, p < 0.001). We applied the final multivariate model obtained for French women in the 2003 ENP to the 1995 database restricted to French women. Variables associated with Toxoplasma infection were the same as those identified using the 2003 sample (Table 2). Although occupational status of couple was not significantly associated with toxoplasmosis prevalence in 1995, prevalence ratios were close to those observed in 2003. The change in prevalence between the two studies differed according to the womens’ age (< 30 years versus age 30 or over) as indicated by a significant interaction term ( p < 0.001) between study year and age. The OR associated with the study year was calculated before and after removing each variable from the saturated multivariate model. Removing each variable did not result in any substantial change (< 10%) of the OR associated with the study year, so the central estimate was independent of other variables associated with toxoplasmosis. We therefore calculated the 2003 to 1995 univariate prevalence ratio to assess change of prevalence over time by age group, which was 0.74 [0.71–0.77] for women under 30 years of age and 0.83 [0.81–0.86] for those aged

We estimated toxoplasmosis seroprevalence on a national level in 1995 (n = 13,459) and 2003 (n = 15,108) using the data collected during repetitive national perinatal surveys. The representation of the 2003 survey was considered satisfactory when compared to the total number of births registered in France in 2003 [19] and the completeness of data collection was good since it ranged by variable from 93.0 to 98.3% [19,20]. However, our study had several limitations that should be taken into account in its interpretation. The toxoplasmosis serology data collected were those produced by routine laboratories which used different techniques and reagents whose sensitivity probably varies. This may, in theory, have an impact on our prevalence estimate by region if testing habits varied by region. We do not have any information on laboratory testing practices between 1995 and 2003 and its potential impact on the change of prevalence over time. In spite of the large sample size, analysis at the district level was difficult due to the small number of mothers residing in some districts. In 8 years, toxoplasmosis prevalence among women of childbearing age dropped almost 20% from 54.3% in 1995 to 43.8% in 2003. A similar decrease has been observed in the USA. The National Health and Nutrition Examination Survey (NHANES) conducted in 1999–2004 showed a 38.8% reduction in T. gondii prevalence among 12- and 49-year-old US-born women since the NHANES III conducted in 1988– 1994 (8.2% for 1999–2004 period) [23]. The drop in the national mutton production-to-consumption ratio from 73% in 1985 to 46% in 2006 observed by the French Livestock Agency (http://www.vienne.chambagri.fr/Reperes/ Filieres/ViandeOvine.htm) associated with the importation of mutton from countries with a low T. gondii prevalence such as the United Kingdom (Ovine-Caprine Industry, France, www.office-elevage.fr) may have contributed to lower exposure to Toxoplasma infection and therefore to lower prevalence of past infection. Deep freezing used for meat transport that kills tissue cysts may also have contributed to decreased human exposure to toxoplasmosis and prevalence of infection. Other measures to control toxoplasmosis in animals based on hygienic practices which prevent contacts between cats and livestock and forage, chemoprophylaxis in late gestation and vaccination of ewes with a live attenuated vaccine (S48 strain) have also played a role [24,25]. Prevalence of toxoplasmosis varied by region of residence in the same way in 1995 and 2003. A link with eating habits or geoclimatic factors such as temperatures, precipitation and altitude has been proposed to explain these regional disparities. In particular, a humid and warm environment favors the survival of oocysts in the soil which may contribute to increased infection rate [7,26]. A similar geographic trend as seen in pregnant women has been documented for the seroprevalence of toxoplasmosis in the ovine flock in France. Surveys conducted among sheep in France between 1960 and 1997

246

F. Berger et al. / Revue d’E´pide´miologie et de Sante´ Publique 57 (2009) 241–248

Table 2 Multivariate analysis of factors associated with Toxoplasma infection in pregnant women in the French National Perinatal Surveys, 1995 and 2003. Pregnant women with French nationality 2003 N = 12,259 Prevalence ratio

CI 95%

p

1995 N = 10,839 Prevalence ratio

CI 95%

< 0,001

Age < 20 20–24 25–29 30–34 35–39 > 39

1.0 1.1 1.3 1.5 1.8 1.9

Number of pregnancies One At least two

1.0 1.1

Education level > Baccalaureate (reference) Last year of high school Secondary school Elementary school

1.0 1.0 1.0 0.8

Region of residence East Center-East West North Paris Basin Mediterranean Southwest Greater Paris Overseas departments

1.0 1.2 1.3 1.4 1.5 1.6 1.7 1.8 1.9

Occupational status of couple Executives Trade services/merchants Intermediate occupations Employees Manual workers No occupation

1.0 1.0 1.0 0.9 1.0 1.0

< 0,001 1.0 1.1 1.2 1.3 1.4 1.5

0.9–1.3 1.1–1.6 1.3–1.8 1.5–2.2 1.6–2.3

0.9–1.3 1.0–1.4 1.1–1.6 1.1–1.6 1.2–1.8

0.04

0.04 1.0 1.1

1.0–1.1

1.0–1.1

0.1

0.2 1.0 1.0 1.0 1.1

0.9–1.0 0.9–1.0 0.7–1.0

0.9–1.0 1.0–1.1 1.0–1.2

< 0.001

< 0.001 1.0 1.1 1.3 1.5 1.5 1.5 1.6 1.7 1.7

1.1–1.6 1.1–1.7 1.3–2.2 1.3–2.2 1.4–2.5 1.5–3.0 1.6–2.0 1.7–2.2

1.0–1.3 1.2–1.4 1.4–1.7 1.3–1.6 1.3–1.6 1.5–1.8 1.6–1.9 1.5–1.9 < 0.001

0.09 0.9–1.1 0.9–1.0 0.9–1.0 0.9–1.0 0.8–1.1

have shown lower prevalence ( 31%) in the Northeast and the West and higher prevalence ( 72%) in Aquitaine, Paris and Coˆte-d’Or (source: Afssa – French Agency for Food Safety). Prevalence among sheep should nevertheless be considered with some caution as surveys were not considered representative. In France, a case control study of toxoplasmosis infection among pregnant women identified eating uncooked or undercooked mutton or beef and eating raw fruits and vegetables outside the home as independent risk factors [27]. A positive correlation between regional consumption of mutton and toxoplasmosis prevalence was documented based on the 1995 ENP [18]. Based on data published by the French Livestock Agency, a similar correlation between consumption of mutton and toxoplasmosis prevalence is also obtained with 2003 data (Spearman’s rank correlation coefficient = 0.8; p = 0.01, Fig. 2). Data published by the French Livestock Agency also show that subjects with a higher socioeconomical status eat more

p

1.0 0.9 0.9 0.9 0.9 0.9

0.8–1.0 0.8–0.9 0.8–0.9 0.8–1.0 0.8–1.0

mutton than others which is consistent with the greater prevalence found in women of higher socioeconomical status. Moreover, mutton consumption in France dropped by 14% between 1999 and 2003 from 5.0 kg per inhabitant per year to 4.3 kg per inhabitant per year (source: French Livestock Agency) as observed for toxoplasmosis seroprevalence. Based on all above data, we believe that mutton consumption is an important determinant of toxoplasmosis infection prevalence variation by region and over time. In the Netherlands and the USA, prevalence of toxoplasmosis rises when education level drops [7,17], which is the inverse of what we observed in France. According to Jones et al. [7], a person of low socioeconomical status may have more contacts with the soil as well as poorer kitchen hygiene which are potential risk factors for Toxoplasma infection. This difference in toxoplasmosis epidemiology by country may be linked to dietary factors: consumption of undercooked mutton or beef plays an important role in France and was not found in the USA.

F. Berger et al. / Revue d’E´pide´miologie et de Sante´ Publique 57 (2009) 241–248

247

References

Fig. 2. Toxoplasmosis prevalence by region defined by Secodip, adjusted for age, according to mutton consumption (index base 100, national average for the year 2003). Regression line: toxoplasmosis prevalence = 0.11  mutton consumption + 31.0 (R2 = 0.39).

Toxoplasmosis prevalence increased with the number of pregnancies, regardless of age. A study conducted in Brazil between 1997 and 1999 suggested that pregnant women were more vulnerable to T. gondii [28]. Moreover the 1995 perinatality study showed a two-fold greater incidence of toxoplasmosis for primipara than for multipara [18]. However, a greater susceptibility for toxoplasmosis during pregnancy does not appear to be sufficient to explain differences between primiparous and multiparous toxoplasmosis seroprevalence. The decrease in toxoplasmosis prevalence between 1995 and 2003 differed by nationality with a smaller drop for women born in North Africa. Although migrants from North Africa tend to adopt eating habits of French residents, they eat less beef and dairy products but more mutton and more fruit, grains and dried vegetables [29]. The decrease between 1995 and 2003 was greater in women under 30 years of age than those aged 30 or over which may indicate that the risk of contracting toxoplasmosis decreased more rapidly among younger women. Toxoplasmosis prevalence decreased among pregnant women, which indicates that the risk of contracting toxoplasmosis is lower than before. This change probably reflects the positive impact of prevention hygiene messages and the changes in mutton supply in France over time. However, prevalence remains greater in France than in the other Northern European countries. If the decrease continues, the relevance of the French current prevention program, which is based on the screening of all pregnant women, will have to be reconsidered. In 2003, more than one out of two women were susceptible of developing Toxoplasma infection during pregnancy. To reduce further the number of cases of CT, it is therefore important to promote measures that reduce the risk during pregnancy (e.g. food habits and hygienic measures). Acknowledgements B. Blondel and K. Supernant of Inserm unit no. 149 for their collaboration, to T. Ancelle for advice and J. Renner of DSDS/ Institut Pasteur de la Guyane for the map.

[1] Dubey JP. Advances in the life cycle of Toxoplasma gondii. Int J Parasitol 1998;28:1019–24. [2] Bowie WR, King AS, Werker DH, Isaac-Renton JL, Bell A, Eng SB, et al. Outbreak of toxoplasmosis associated with municipal drinking water. Lancet 1997;350:173–7. [3] Benenson MW, Takafuji ET, Lemon SM, Greenup RL, Sulzer AJ. Oocysttransmitted toxoplasmosis associated with ingestion of contaminated water. N Engl J Med 1982;307:666–9. [4] Toxoplasmose : e´tat des connaissances et e´valuation du risque lie´ a` l’alimentation. Rapport du groupe de travail ‘‘Toxoplasma gondii’’. 2005. Available on http://www.afssa.fr/Documents/MIC-Ra-Toxoplasmose.pdf (site consulte´ le 18 juin 2009). [5] Desmonts G, Couvreur J, Ben Rachid MS. Le toxoplasme, la me`re et l’enfant. Arch Fr Pediatr 1965;22:1183–200. [6] Papoz L, Sarmini H, Funes A, Comiti V, et al. E´tude de la pre´valence de l’empreinte immunologique de la rube´ole, de la toxoplasmose, du cytome´galovirus, de l’herpe`s et de l’he´patite B chez 8594 femmes de 15 a` 45 ans en France en 1982–1983. Bull Epidemiol Hebd 1984;20:2–3. [7] Jones JL, Kruszon-Moran D, Wilson M, McQuillan G, Navin T, McAuley JB. Toxoplasma gondii infection in the United States: seroprevalence and risk factors. Am J Epidemiol 2001;154:357–65. [8] 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 U.K.. J Infect 1998;36:189–96. [9] Petersson K, Stray-Pedersen B, Malm G, Forsgren M, Evengard B. Seroprevalence of Toxoplasma gondii among pregnant women in Sweden. Acta Obstet Gynecol Scand 2000;79:824–9. [10] Nissapatorn V, Noor Azmi MA, Cho SM, Fong MY, Init I, Rohela M, et al. Toxoplasmosis: prevalence and risk factors. J Obstet Gynaecol 2003;23: 618–24. [11] Bouratbine A, Siala E, Chahed MK, Aoun K, Ben Ismail R. Seroepidemiologic profile of toxoplasmosis in Northern Tunisia. Parasite 2001;8:61–6. [12] el Nawawy A, Soliman AT, el Azzouni O, Amer e, Karim MA, Demian S, et al. Maternal and neonatal prevalence of Toxoplasma and cytomegalovirus (CMV) antibodies and hepatitis-B antigens in an Egyptian rural area. J Trop Pediatr 1996;42:154–7. [13] Faye O, Leye A, Dieng Y, Richard-Lenoble D, Diallo S. La toxoplasmose a` Dakar. Sondage se´roe´pide´miologique chez 353 femmes en aˆge de procre´er. Bull Soc Pathol Exot 1998;91:249–50. [14] Guebre-Xabier M, Nurilign A, Gebre-Hiwot A, Hailu A, Sissay Y, Getachew E, et al. Sero-epidemiological survey of Toxoplasma gondii infection in Ethiopia. Ethiop Med J 1993;31:201–8. [15] Diaz-Suarez O, Estevez J, Garcia M, Cheng-Ng R, Araujo J, Garcia M. Seroepidemiology of toxoplasmosis in a Yucpa Amerindian community of Sierra de Perija, Zulia State, Venezuela. Rev Med Chil 2003;131: 1003–10. [16] Fuente MC, Bovone NS, Cabral GE. Prophylaxis of prenatal toxoplasmosis. Medicina (B Aires) 1997;57:155–60. [17] Kortbeek LM, De Melker HE, Veldhuijzen IK, Conyn-Van Spaendonck MA. Population-based Toxoplasma seroprevalence study in The Netherlands. Epidemiol Infect 2004;132:839–45. [18] Ancelle T. La toxoplasmose chez la femme enceinte en France en 1995. Re´sultats d’une enqueˆte nationale pe´rinatale. RNSP; 1995. Available on http://www.bdsp.ehesp.fr/base/Scripts/ShowA.bs?bqRef=143624 (site consulte´ le 18 juin 2009). [19] Blondel B, Supernant K, du Mazaubrun C, Breart G. Enqueˆte nationale pe´rinatale 2003. Situation en 2003 et e´volution depuis 1998. 2005. Available on http://www.sante.gouv.fr/htm/dossiers/perinat03/enquete. pdf (site consulte´ le 18 juin 2009). [20] Blondel B, Supernant K, du Mazaubrun C, Breart G. La sante´ pe´rinatale en France me´tropolitaine de 1995 a` 2003 : Re´sultats des enqueˆtes nationales pe´rinatales. J Gynecol Obstet Biol Reprod (Paris) 2006;35: 373–87. [21] Zou G. A modified Poisson regression approach to prospective studies with binary data. Am J Epidemiol 2004;159:702–6.

248

F. Berger et al. / Revue d’E´pide´miologie et de Sante´ Publique 57 (2009) 241–248

[22] Greenland S. Model-based estimation of relative risks and other epidemiologic in studies of common outcomes and in cas-control studies. Am J Epidemiol 2004;160:301–5. [23] Jones JL, Kruszon-Moran D, Sanders-Lewis K, Wilson M. Toxoplasma gondii infection in the United States, 1999–2004. Decline from the prior decade. Am J Trop Med Hyg 2007;77:405–10. [24] Afssa. Rapport du groupe de travail « alimentation animale et se´curite´ sanitaire des aliments ». Afssa; 2000. Available on http://lesrapports. ladocumentationfrancaise.fr/BRP/004001815/0000.pdf (site consulte´ le 18 juin 2009). [25] Chartier C, Mallereau M-P. E´fficacite´ vaccinale de la souche S48 de Toxoplasma gondii vis-a`-vis d’une infection expe´rimentale chez la che`vre. Ann Med Ve 2001;145:202–9.

[26] Barbier D, Ancelle T, Martin-Bouyer G. Seroepidemiological survey of toxoplasmosis in La Guadeloupe. French West Indies. Am J Trop Med Hyg 1983;32:935–42. [27] Baril L, Ancelle T, Goulet V, Thulliez P, Tirard-Fleury V, Carme B. Risk factors for Toxoplasma infection in pregnancy: a case-control study in France. Scand J Infect Dis 1999;31:305–9. [28] Avelino MM, Campos Jr D, Parada JB, Castro AM. Risk factors for Toxoplasma gondii infection in women of childbearing age. Braz J Infect Dis 2004;8:164–74. [29] HCSP. Pour une politique nutritionnelle de sante´ publique en France. HCSP; 2000. Availble on http://www.hcsp.fr/hcspi/docspdf/hcsp/hc000933.pdf (site consulte´ le 18 juin 2009).