- Email: [email protected]
Prevalence of toxoplasmosis in France in 1998: Is there a difference between men and women? At what age do children become infected?
Available online at
www.sciencedirect.com Revue d’E´pide´miologie et de Sante´ Publique 61 (2013) 311–317
Original article
Prevalence of toxoplasmosis in France in 1998: Is there a difference between men and women? At what age do children become infected? La pre´valence de la toxoplasmose en France en 1998 est-elle diffe´rente entre les hommes et les femmes ? A` quel aˆge s’infectent les enfants ? H. Bellali a,b,*, H. Pelloux c, I. Villena d, H. Fricker-Hidalgo c, Y. Le Strat a, V. Goulet a,* a
De´partement des maladies infectieuses, institut de veille Sanitaire, 12, rue Val-d’Osne, 94415 Saint-Maurice cedex, France b Profet – Programme de formation a` l’e´pide´miologie de terrain, 94415 Saint-Maurice cedex, France c Laboratoire de parasitologie-mycologie, universite´ Joseph-Fourier, centre hospitalier universitaire de Grenoble, 38000 Grenoble, France d Centre national de re´fe´rence de la toxoplasmose, universite´ de Reims – Champagne-Ardenne, SFR Cap-Sante´, centre hospitalier universitaire de Reims, 51100 Reims, France Received 10 January 2012; accepted 15 November 2012
Abstract Background. – The only national seroprevalence data currently available on toxoplasmosis in France are from the national perinatal surveys of pregnant women conducted in 1995 and 2003. These surveys are national, exhaustive and cross-sectional studies of all pregnant women who give birth in France during one specified week. These cross-sectional studies, conducted among women of childbearing age (defined as 18 to 45 years), showed a positive correlation between seroprevalence and age, with a significant regional disparity. This study was performed in order to compare the prevalence of toxoplasmosis antibodies in men and women in the 18–45 age group, to confirm regional variations and to estimate the seroprevalence of toxoplasmosis in France for different age groups, particularly among children and among adults aged over 45 years. Methods. – Serum samples from 2060 subjects were available from a national serum bank that was established in 1997 as part of a European study on vaccine preventable diseases. The sera were tested for IgG antibodies in 2008–2009, by ELISA test, at the laboratory of parasitologymycology, CHU Grenoble. Results. – The seroprevalence for the population aged 1–64 years was 55.4%. Seroprevalence did not vary between the sexes, except among those aged over 45 years, where it was higher in men than in women. Toxoplasmosis seroprevalence varied significantly by regions for all ages. It increased with age and we noted a stronger increase in prevalence in adolescents (10–20 years) than in other age groups. Conclusion. – This study showed that children have limited exposure to Toxoplasma gondii and that seroprevalence in men and women does not differ for the population aged 45 years and under. This study confirms the geographical disparity in prevalence in France that has been found in other studies in women of childbearing age. This disparity cannot be explained by different laboratory techniques, because sera were tested in a single laboratory. The study also raises the possibility of extrapolating seroprevalences from ENP to the general population and thus estimating the seroprevalence in the French population. # 2013 Published by Elsevier Masson SAS. Keywords: Ttoxoplasmosis; Antibodies; Prevalence; France
Re´sume´ Position du proble`me. – Les seules donne´es nationales de la se´ropre´valence de la toxoplasmose disponibles en France a` ce jour proviennent des enqueˆtes nationales pe´rinatales re´alise´es en 1995 et 2003 chez la femme enceinte. Ces enqueˆtes nationales, qui sont des e´tudes transversales exhaustives re´alise´es pour toutes les naissances en France durant une semaine, ont montre´ une corre´lation positive de la se´ropre´valence de la toxoplasmose avec l’aˆge et une disparite´ re´gionale importante. Les objectifs de ce travail e´taient de comparer la pre´valence entre les hommes et les femmes en aˆge de procre´er (18–45 ans), de ve´rifier les variations entre les re´gions et d’estimer la se´ropre´valence de la toxoplasmose en France selon l’aˆge, notamment chez les enfants et les adultes aˆge´s de plus de 45 ans.
* Corresponding author. E-mail address: [email protected], [email protected] (H. Bellali), [email protected] (V. Goulet). 0398-7620/$ – see front matter # 2013 Published by Elsevier Masson SAS. http://dx.doi.org/10.1016/j.respe.2012.11.005
312
H. Bellali et al. / Revue d’E´pide´miologie et de Sante´ Publique 61 (2013) 311–317
Me´thodes. – Un e´chantillon de se´rum concernant 2060 sujets provenait d’une se´rothe`que nationale mise en place dans le cadre d’une e´tude se´roe´pide´miologique europe´enne re´alise´e en 1997 pour e´tudier les maladies a` pre´vention vaccinale. La recherche des anticorps antitoxoplasmose type IgG a e´te´ faite graˆce au test Elisa en 2008–2009 au laboratoire de Parasitologie-Mycologie, CHU de Grenoble. Re´sultats. – La se´ropre´valence de la toxoplasmose chez la population aˆge´e de un a` 64 ans e´tait de 55,4 %. Elle n’e´tait pas diffe´rente entre les hommes et les femmes a` l’exception de ceux qui sont aˆge´s de plus de 45 ans ou` elle e´tait plus e´leve´e chez les hommes. Elle variait significativement selon les re´gions et elle augmentait avec l’aˆge, avec une augmentation importante a` l’adolescence. Conclusion. – Cette e´tude montre que les enfants sont peu expose´s a` la toxoplasmose et que jusqu’a` l’aˆge de 45 ans la se´ropre´valence des hommes n’est pas significativement diffe´rente de celle des femmes. Cette e´tude confirme que la disparite´ ge´ographique de la se´ropre´valence de la toxoplasmose en France de´ja` de´crite ne peut pas eˆtre explique´e par des diffe´rences de techniques entre laboratoires car tous les se´rums ont e´te´ teste´s en un seul lieu. Elle permet de conside´rer que les se´ropre´valences de l’Enqueˆte nationale pe´rinatale peuvent eˆtre extrapole´es aux hommes et donnent des estimations de la pre´valence de la toxoplasmose dans la population ge´ne´rale. # 2013 Publie´ par Elsevier Masson SAS. Mots cle´s : Toxoplasmose ; Anticorps ; Pre´valence ; France
1. Introduction Toxoplasmosis is a zoonotic disease caused by the protozoan parasite Toxoplasma gondii [1]. This infection is prevalent in humans and animals worldwide [2]. Felids are the key animal species in the life cycle of this parasite because they are the hosts that can excrete the environmentally-resistant stage, the oocyst [3]. The prevalence of T. gondii varies greatly in populations throughout the world and seems to be affected by food-production practices, water treatment, climate, topography, soil and cat feces-related hygiene, occupational and nonoccupational soil exposure, and culinary practices. The infection is more prevalent in some regions in Europe and parts of South America than in Asia, the United States or Australia [4–6]. In humans, infection is usually either asymptomatic or the cause of mild flu-like symptoms. However, toxoplasmosis can be life threatening in immunocompromised individuals. Moreover, if acquired during pregnancy, toxoplasmosis can lead to a wide variety of manifestations in the fetus and infant, including spontaneous abortion, stillbirth, or a live infant with classic signs of congenital toxoplasmosis such as hydrocephalus or microcephalus, cerebral calcifications and retinochoroiditis [2]. A prevention program for congenital toxoplasmosis was implemented in France in 1978. Systematic serological screening of all pregnant women has been undertaken since 1985. Monthly monitoring was implemented in 1992. It involves follow-ups of toxoplasmosis serology for all non-immune pregnant women. These women receive primary prevention advice on food hygiene, food preparation and avoiding eating certain types of foods [7]. Diagnosis of maternal seroconversion during pregnancy is necessary to prevent transmission to the fetus, by means of specific treatment and also to permit close monitoring of at-risk fetuses and neonates. A surveillance program for congenital toxoplasmosis has been operating in France since 2006, and is coordinated by the National Reference Center for toxoplasmosis [8]. The only current national seroprevalence data available in France are from the national perinatal surveys (ENP) conducted in 1995 and 2003 [9,10]. ENP surveys are national, exhaustive and cross-sectional studies of all pregnant women who give birth in France during one specified week. These studies,
conducted among women of childbearing age (defined as 18 to 45 years), showed a positive correlation between seroprevalence and age, with a significant regional disparity. The main objectives of this study were to compare the prevalence of toxoplasmosis antibodies between men and women in the 18–45 age group in order to establish whether the results from national perinatal surveys (ENP) could be extrapolated to the general population and thus estimate the prevalence in the French population, and to check geographic variations. The second objective was to estimate toxoplasmosis seroprevalence according to age, mainly among children and adults over the age of 45.
2. Material and methods In 1997, a cross-sectional study was carried out on sera collected from a sample of the French population within the framework of the European sero-epidemiology network (ESEN), which coordinates and harmonizes serological surveillance of immunity to vaccine preventable diseases. National banks of serum specimens have been collected in six European countries (Denmark, England, France, Germany, Italy and the Netherlands) to provide about 3500 specimens for each country, stratified by age, with males and females equally represented [11]. In France, specimens were collected at 39 volunteer centers (30 private biomedical laboratories, three hospital biomedical laboratories and six hospital departments), between January 1998 and July 1999. Subjects were recruited by biomedical laboratories from people who came to the laboratory for a blood sample for any reason. After consent, additional tubes of blood were collected for the purposes of the study. During the sampling, the biologist collected information from each individual on sex, age, administrative district of residence, country of birth, health insurance coverage, and month and year of sampling. For adults, education level and occupational category were recorded, and for children (< 18 years), the occupational category of the head of household. Each laboratory had a quota of persons by age and sex to sample. Subjects with conditions that could affect their immunity were excluded from the ESEN survey. These conditions included being immunocompromised; having received a blood
H. Bellali et al. / Revue d’E´pide´miologie et de Sante´ Publique 61 (2013) 311–317
transfusion or having been treated with gamma globulin or other blood products in the previous 6 months; and having hepatic failure, renal failure, nephrotic syndrome, hematologic diseases or cancers, malnutrition or autoimmune diseases. The final sample underrepresented people living in rural areas and over-represented people in the highest sociooccupational categories. The conserved sera were tested blind for toxoplasmosis antibodies in 2008–2009 (Fig. 1). Serological tests for toxoplasmosis were performed by the laboratory of parasitology-mycology, CHU Grenoble, using ELISA Vidas IgG-II (bioMe´rieux, Marcy l’Etoile, France) and AxSYM IgG (Abbott Laboratories, Wiesbaden, Germany) [12,13]. When the results obtained using the two ELISA techniques were discordant, an IgG in-house immunofluorescence [14] was performed to make a decision on the presence or absence of antibodies against T. gondii. Data were analyzed using STATA version 9.2. Age was grouped into six classes for univariate analysis: 1–9 years, 10– 19 years, 20–29 years, 30–39 years, 40–49 years and 50–64 years. For multivariate analysis, age was introduced as a continuous variable modeled by a fractional polynomial. To take into account the interaction between age and sex, three categories (< 18 years, 18–45 years and 46–64 years) were used to present stratified results of multivariate analysis. The study did not include French Overseas districts. France is divided into eight major areas of territory study and planning (Zones d’e´tudes et d’ame´nagement du territoire [ZEAT]), using the territorial division created in 1967 by the national institute of statistics and economic studies (Institut national de la statistique et des e´tudes e´conomiques [INSEE]). Each ZEAT contains between one and six administrative regions. This division provides areas that are fairly homogeneous, with a sufficient number of subjects to test for differences in seroprevalence. Areas of residence were coded into one of three categories: rural area, which was defined as a town with a population under 2000; urban area with fewer than 200 000 inhabitants; and urban area with 200 000 or more inhabitants, including Greater Paris.
[(Fig._1)TD$IG]
Seroepidemiological survey 1998: 2937 sera collected 688 not available (insufficient amount)
Residual sera: 2249 sera Excluding those born outside France, n=185 And those <1 year, n=4
Study sample: 2060 subjects Fig. 1. Source and number of sera used in toxoplasmosis study, France 1998.
313
In 1998, the prevalence in France for the population aged between 1 and 64 years was estimated by applying the observed prevalence in our study to the French population aged 1–64 years (data INSEE 1998), stratified by age, region and, for those aged 46 years and over, by sex. For each ZEAT, age-specific prevalence rates (by 5 year age groups) and age- and sexspecific rates (for those aged over 46 years) were applied to ZEAT population and summed. The total number of seropositives was then divided by the 1998 total French population aged 1–64 years, to calculate the overall seroprevalence. To compare the results of this study with those of the National Perinatal Survey (ENP) 1995, prevalence was estimated for women aged between 18 and 45 years, by standardizing for age and region of the survey sample 1995 (ENP). We studied the association of toxoplasmosis with the following variables: age, sex, occupational category, ZEAT region and area of residence. Comparison of seroprevalence by age, sex, region of residence, occupation and residential area, was performed first by univariate analysis, then by a multivariate backward logistic regression model and testing possible interactions between all the independent variables. All variables studied were included in the model for multivariate analysis. This choice was justified by the low number of factors studied. Age was introduced as a continuous variable and was modeled using fractional polynomial, giving a relationship between toxoplasmosis seroprevalence and age [15]. 3. Results A total of 2060 sera were tested for IgG toxoplasma antibodies. Data were missing for between four and 96 individuals, depending on the variable. The average age was 24 years (range: 1–64 years, median: 20 years), and men accounted for 49% of subjects in the sample. In 1998, the prevalence of toxoplasmosis in France for the population aged between 1 and 64 years was 55.4% (95% CI: 53.4–57.4%). Prevalence for women aged between 18 and 45 years, after standardizing for age and region of the sample population 1995 of the ENP, was estimated at 55.7% (95% CI: 54.7,56.7%). Table 1 shows population and toxoplasmosis seroprevalence by sociodemographic characteristics. Toxoplasmosis seroprevalence increased with age. It was 4% in children under 10 years and 88% in people aged 50 and over (P < 0.001). Between the ages of 10–19 years, seroprevalence was 7.5 times greater than that observed in children aged less than 10 years. The prevalence of toxoplasmosis varied significantly between rural (31%) and urban (47%) areas (P < 0.001). It did not differ between men and women (45.3% and 41.6% respectively, P = 0.081). Multivariate analysis results are presented separately by age category because of a significant (P = 0.023) interaction between age as a continuous variable and sex (Table 2). Toxoplasmosis seroprevalence varied significantly by area whatever the age. A low prevalence was observed in the
314
H. Bellali et al. / Revue d’E´pide´miologie et de Sante´ Publique 61 (2013) 311–317
Table 1 Study population distribution and seroprevalence of toxoplasmosis by the factors studied, cross-sectional study, France 1998. na
Seroprevalence (n = 896)
Crude odds ratio (OR)
438 560 406 191 185 278
3.9 30 49.5 61.8 79.5 88.1
Reference 10.1 23.1 38.1 91.1 174.9
Sex Male Female
1006 1048
45.3 41.6
1.2 Reference
Occupational category Craftsman/farmer Manual/non-manual workers Professional/executive Other/inactive Student Retired
87 1025 348 231 185 115
16.1 47.1 21.3 47.2 49.7 79.1
Reference 4.6 1.4 4.6 5.1 19.8
2.6–8.4 0.7–2.6 2.4–9.0 2.6–10.1 7.8–50.3
270 341 108 222 277 233 383 225
29.6 43.4 47.2 32.4 39.7 48.1 54 51.1
Reference 1.8 2.1 1.1 1.6 2.2 2.8 2.5
1.3–2.6 1.3–3.4 0.8–1.7 1.1–2.2 1.5–3.2 2.0–3.9 1.7–3.6
212 430 1322
31.6 35.3 47.6
Reference 1.2 1.9
0.8–1.7 1.4–2.7
Age (years) 1–9 10–19 20–29 30–39 40–49 50–64
ZEAT regions Centre-East Paris Basin North East West South-West Greater Paris Mediterranean region Residential area Rural (< 2000 inhabitants) Urban (< 200 000 inhabitants) Urban ( 200 000 inhabitants and Greater Paris) a
95% CI
P
< 0.001 5.9–17.2 12.7–41.9 18.4–78.7 35.3–234.9 56.4–541.8 0.9–1.4
0.081
< 0.001
< 0.001
< 0.001
Data were missing for between 1 (ZEAT regions) and 96 (Residential area) individuals, depending on the variable.
Centre-East (30.7%), a high prevalence in Greater Paris (57.5%) and the South-West (52.3%) and medium prevalence (35–50%) in other areas (East, West, Paris Basin, North and the Mediterranean Region) (Fig. 2). In children (< 18 years) and in adults (18–45 years), toxoplasmosis prevalence increased with increasing age, but did not vary between sexes. Over 45 years, a significantly higher prevalence was observed in men (OR = 2, P = 0.033) than in women. In multivariate analysis, once the interaction between sex and age was taking into account, neither occupational category nor living in a rural or urban district was associated with toxoplasmosis. In order to represent the prevalence age profile, we calculated estimated prevalence for each year of age and among men and women. We represented the obtained curve with observed data (Fig. 3). 4. Discussion In 1998, the prevalence of toxoplasmosis in France in the population aged between 1 and 64 years was estimated at 54%. Toxoplasmosis seroprevalence increased with age, with a more
marked rise during adolescence. Prevalence did not differ between men and women between the ages of 18 to 45 years but was higher in men over 45 years than in women in the same age group. These sera were collected from individuals attending medical laboratories, so were not a random sample of the population designed for a French seroepidemiological survey for vaccine preventable diseases. This sample underestimated people living in rural areas and over-represented people in the highest socio-occupational categories. These two factors were not, however, associated with toxoplasmosis prevalence in multivariate analysis. The distribution of these two factors was similar between men women. Seroprevalence among women aged 18–45 years was estimated at 55.7%, which is similar to the seroprevalence observed in the 1995 ENP study (54.3%) [9]. The highest prevalences were observed in Greater Paris. The South-West and Centre-East regions of France had the lowest prevalence. National perinatal surveys conducted in 1995 [9] and 2003 [10] showed the same regions of high and low prevalence. The results of toxoplasmosis serology collected in national perinatal studies were produced by routine laboratories, which used
H. Bellali et al. / Revue d’E´pide´miologie et de Sante´ Publique 61 (2013) 311–317
315
Table 2 Risk factors of toxoplasmosis seroprevalence in France, 1998: results of multivariate analysis by age category. Age category 1–17 years (n = 869) Variables
OR
95% CI
Age (in years) 1–4 5–9 10–14 15–17 18–24 25–29 30–34 35–39 40–45 46–49 50–54 55–59 60–64
1 3.1 9.6 23.6 – – – – – – – – –
2.3–3.9 4.5–14.6 6.2–41.0 – – – – – – – – –
Sex Male Female
0.98 1
ZEAT regions Centre-East Paris Basin North East West South–West Greater Paris Mediterranean region
1 2.5 0.9 1.3 1.4 3.6 3.9 3.8
18–45 years (n = 838)
46–64 years (n = 350)
P-value
OR
95% CI
P-value
OR
95% CI
P-value
< 0.001 0.170 < 0.001 < 0.001 –
– – –
– – – –
–
– – – – – – – – – 1 0.98 0.95 0.93
– – – – – – – – –
–
0.7–1.3 0.4–1.5 0.1–1.8
0.875 0.088 0.187 0.881
1.1–4.0
0.033
2.3–25.0 1.1–15.9 0.7–6.1 1.2–8.8 2.2–140.4 2.8–29.2 2.2–30.6
< 0.001 0.001 0.04 0.2 0.022 0.007 < 0.001 0.002
–
1 1.4 1.9 2.5 3.2 – – – –
0.6–1.5
0.9
1.3 1
1.0–6.1 0.2–4.7 0.5–3.6 0.5–3.8 1.4–9.0 1.6–9.5 1.5–9.5
< 0.001 0.037 0.902 0.626 0.468 0.006 0.002 0.005
1 2.8 2.17 1.9 2.2 3.7 5.1 2.93
1.3–1.6 1.5–2.3 1.7–3.2 1.9–4.5 – – – –
< 0.001 0.080 0.002 0.034 < 0.001 –
0.9–1.7
0.08
2 1
1.7–5.0 1.2–4.6 1.1–3.4 1.3–3.8 2.1–6.6 3.1–9.0 1.6–5.1
< 0.001 < 0.001 0.017 0.033 0.005 < 0.001 < 0.001 < 0.001
1 7.6 4.1 2.1 3.2 17.4 9 8.1
(Fig._2)TD$IG][OR: adjusted odds ratio.
Fig. 2. Seroprevalence of toxoplasmosis by area, standardized for age and sex, general population, France, 1998.
–
[(Fig._3)TD$IG] 316
H. Bellali et al. / Revue d’E´pide´miologie et de Sante´ Publique 61 (2013) 311–317 1 0,9
Seroprevalence
0,8 0,7 0,6 0,5 0,4 0,3
Observed prevalence
0,2
Estimated prevalence among women
Estimated prevalence
Estimated prevalence among men
0,1 0 0
10
20
30
40
50
60
70
Age (years)
Fig. 3. Estimated prevalence of toxoplasmosis by age and sex, France 1998. h i h i Equation for men: prevalenceðageÞ ¼ exp 7:79 ððage þ 1Þ=10Þ1=2 0:619 = 1 þ exp 7:79 ððage þ 1Þ=10Þ1=2 0:619 Equation for women: prevalenceðageÞ ¼ expð2:01 ½Inððage þ 1Þ=10Þ 0:918Þ=ð1 þ expð2:01 ½Inððage þ 1ÞÞ=10 0:918ÞÞ.
different techniques and reagents with potentially variable sensitivities. If testing habits did vary between regions, this would have an impact on the prevalence found in the study. In our study, regional disparities cannot be explained by different laboratory techniques because the sera were tested in the same laboratory. This local variation is most likely to be related to culinary practices and the ability of oocysts to survive in different climates. A positive correlation between regional consumption of mutton and toxoplasmosis prevalence has been shown by Ancelle et al. in ENP 1995 [9] and Berger et al. in ENP 2003 [10]. Halos et al. [16] showed that districts which had high seroprevalence in sheep were the same districts that had high seroprevalence in women of childbearing age in ENP surveys. This observation could be explained by an intense circulation of the parasite in these areas and may be due to specific environmental conditions (such as density of felids, climate characteristics and dietary habits). Interestingly, the geographical disparities are observed from childhood with a non-significant increase with age. This study is important because to our knowledge it is the first in France to estimate and compare the prevalence of toxoplasmosis in males and females, and in children and in the elderly. Seroprevalence was very low in children under 10 years old (4%). This low prevalence was also reported in a study conducted recently in a rural region of France (2.5% estimated seroprevalence in 7-year-olds) [17]. In light of these results, the transmission of toxoplasmosis from soil contaminated by cat feces does not seem to be the major route of contamination in France, since contact with the ground starts in early childhood (3–5 years). This is the first French study to provide prevalence data for men, because most previous studies have investigated toxoplasmosis in pregnant women. In general, prevalence does not vary greatly between the sexes [18–24]. In our study, seroprevalence did not vary between the sexes, except in men and women over 45 years old, where it was higher in men. It is known that the prevalence of T. gondii antibodies increases with age, as has been found by several studies [5,9,10,17,18,25–30]. This reflects the increased probability of having been in contact with the parasite through one of the transmission routes as people get older.
We noted an important increase in prevalence during adolescence. The same finding has been observed in many other studies [19,25,27–29]. Fernandes et al. [25] estimated the average age of acquisition of toxoplasmosis at 11 years, through mathematical modeling based on a seroepidemiological study in a Brazilian population between 1984 and 2006. Taylor et al. [19] found a nearly linear increase in seroprevalence between 10 and 14 years of age in urban and rural schoolchildren aged 4 to 18 years in Dublin, Wicklow and Kildare (Ireland). In Doha (Qatar), Abu-Madi [27] observed a prevalence that increased from 6% in the 2–10 year age group to 16% in the 11–20 year age group. SalahiMoghaddam [28] reported an increase in toxoplasmosis prevalence of almost 30% from the 5–9 year age group to the 10–14 year age group. In a community-based survey of human toxoplasmosis in rural Amazonia [29], the log-probit model fitted to age-prevalence data estimated the median age of seroconversion at 14 years. These observations may be explained by a change in eating habits of children of this age, with a higher consumption of meat including mutton or lamb, known to be a main source of transmission of Toxoplasma worldwide and especially in Europe [31,32]. Moreover, studies of consumption of meat in France (INCA 1999 and CCAF 2003) have shown a steady increase in the average daily consumption of meat with age and a tendency in young people (15–24 years) to prefer eating red meat (beef and lamb) [33]. The study reported here allowed us to estimate changes in seroprevalence by age between 1 and 64 years. A similar age profile of estimated probability of seropositivity was described by Fromont et al. [17] in a toxoplasma seroprevalence study in the French rural population in 2004 and also by Ferreira et al. [29] in a communitybased survey of human toxoplasmosis in rural Amazonia, Brazil. Prevalence did not differ between men and women between the ages of 18 to 45 years. These results show that the estimated seroprevalence surveys in ENP can be extrapolated to the general population using the estimated curve of seroprevalence by age.
H. Bellali et al. / Revue d’E´pide´miologie et de Sante´ Publique 61 (2013) 311–317
Disclosure of interest The authors declare that they have no conflicts of interest concerning this article.
[17]
[18]
References [1] Ryan KJ, Ray CG, Sherris JC. Toxoplasmosis. In: Sherris medical microbiology: an introduction to infectious diseases4th ed., United States: McGraw Hill; 2004. p. 723–7. [2] Montoya JG, Liesenfeld O. Toxoplasmosis. Lancet 2004;363:1965–76. [3] Hutchison WM, Dunachie JF, Siim JC. Cats as a source of toxoplasmosis. Proc R Soc Med 1972;65:1001. [4] McQuillan G, Kruszon-Moran D, Kottiri BJ, Curtin LR, Lucas JW, Kington RS. Racial and ethnic differences in the seroprevalence of 6 infectious diseases in the United States: data from NHANES III, 1988– 1994. Am J Public Health 2004;94:1952–8. [5] Jones JL, Kruszon-Moran D, Wilson M, McQuillan G, Navin T, McAuley JB. Toxoplasmosis in the United States: seroprevalence and risk factors. Am J Epidemiol 2001;154:357–65. [6] Jones JL, Dargelas V, Roberts J, Press C, Remington JS, Montoya JG. Risk factors for Toxoplasma gondii infection in the United States. Clin Infect Dis 2009;49:878–84. [7] Carme B, Tirard-Fleury V. La toxoplasmose chez la femme enceinte en France : se´ropre´valence, taux de se´roconversion et niveau de connaissance des mesures pre´ventives. Tendances 1965–1995. Med Mal Infect 1996;26:431–6. [8] King L, Villena I, Ancelle T, Wallon M, Thulliez P, et al. La toxoplasmose conge´nitale : mise en place d’un dispositif de surveillance en France. Bull Epidemiol Hebdomadaire 2008;14-15:122–4. [9] Ancelle T, Goulet V, Tirard-Fleury V, Baril L, Mazaubrun C, Thulliez P, et al. La toxoplasmose chez la femme enceinte en France en 1995. Re´sultats d’une enqueˆte nationale pe´rinatale. Bull Epidemiol Hebdomadaire 1996;51:227–9. [10] Berger F, Goulet V, Le strat Y, Desenclos JC. Toxoplasmosis among pregnant women in France: risk factors and change of prevalence between 1995 and 2003. Rev Epidemiol Sante Publique 2009;57:241–8. [11] Osborne K, Weinberg J, Miller E. The European sero-epidemiology network. Eurosurveillance 1997;2:29–31. [12] Fricker-Hidalgo H, Saddoux C, Suchel-Jambon AS, Romand S, Foussadier A, Pelloux H, et al. New VIDAS assay for toxoplasma-specific IgG avidity: evaluation on 603 sera. Diagn Microbiol Infect Dis 2006;56:167–72. [13] Gay-Andrieu F, Fricker-Hidalgo H, Sickinger E, Espern A, BernierPinchart MP, Braun HB, et al. Comparative evaluation of the architect toxo IgG, IgM avidity assays for anti-toxoplasma antibodies detection in pregnant women sera. Diagn Microbiol Infect Dis 2009;65:279–87. [14] Roberts A, Hedman K, Luyasu V, Zufferey J, Bessie`res MH, Blatz RM, et al. Multicenter evaluation of strategies for serodiagnosis of primary infection with Toxoplasma gondii. Eur J Microbiol Infect Dis 2001;20:467–74. [15] Royston P, Ambler G, Sauerbrei W. The use of fractional polynomials to model continuous risk variables in epidemiology. Int J Epidemiol 1999;28:964–74. [16] Halos L, The´bault A, Aubert D, Thomas M, Perret C, Geers R, et al. An innovative survey underlining the significant level of contamination by
[19]
[20]
[21]
[22]
[23]
[24]
[25]
[26]
[27]
[28]
[29]
[30]
[31]
[32] [33]
317
Toxoplasma gondii of ovine meat consumed in France. Int J Parasitol 2010;40:193–200. Fromont EG, Riche B, Rabilloud M. Toxoplasma seroprevalence in a rural population in France: detection of a household effect. BMC Infect Dis 2009;9:76–82. 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. Taylor MRH, Lennon b, Holland CV, Cafferkey M. Community study of toxoplasma antibodies in urban and rural schoolchildren aged 4 to 18 years. Arch Dis Child 1997;77:406–9. Alvarado-Esquivel C, Mercado-Suarez MF, Rodriguez-Briones A, FelladTaures L, Alaya-Alaya JO, Nevarez-Piedra LJ, et al. Sero-epidemiology of infection with Toxoplasma gondii in healthy blood donors of Durango, Mexico. BMC Infect Dis 2007;7:75–82. Kook J, Lee HJ, Kim B, Yun ChK, Guk SM, Seo Min. et al. Toxoplasma gondii antibody titers in sera of children admitted to the Seoul National University Children’s Hospital. Korean J Parasitol 1999;37:27–32. Lee YH, Noh HJ, Hwang SK, Shin DW. Seroepidemiological study of Toxoplasma gondii infection in the rural area Okcheon-gun, Korea. Korean J Parasitol 2000;38:251–6. Shin DW, Cha DY, Hua QJ, Cha GH, Lee YH. Seroprevalence of Toxoplasma gondii infection and characteristics of seropositive patients in general hospitals in Daejeon, Korea. Korean J Parasitol 2009;47: 125–30. Yang HJ, Park YK, Hong SC, Bae GM, Lee SH. Seroprevalence of toxoplasmosis in the residents of Cheju Island, Korea. Korean J Parasitol 2000;38:91–3. Fernandes GCVR, Azevedo RS, Amaku M, Yu ALF, Massad E. Seroepidemiology of toxoplasma infection in a metrpolitan region of Brazil. Epidemiol Infect 2009;137:1809–15. Ertug S, Okyay P, Turkmen M, Yuksel H. Seroprevalence and risk factors for toxoplamsa infection among pregnant women in Aydin province, Turkey. BMC Public Health 2005;5:66–71. Abu-Madi M, Al-Molawi N, Behnke JM. Seroprevalence and epidemiological correlates of Toxoplasma gondii infections among patients referred for hospital-based serological testing in Doha, Qatar. Parasites Vectors 2008;1:39–47. Salahi-Moghaddam A, Hafizi A. A serological study on Toxoplasma gondii infection among people in south of Tahran, Iran. Korean J Parasitol 2009;47:61–3. Ferreira MU, Hiramoto RM, Aureliano DP, Da Silva-Nunes M, Da Silva NS, Malafonte RS, et al. A community based survey of human toxoplasmosis in rural Amazonia: seroprevalence, seroconversion rate and associated risk factors. Am J Trop Med Hyg 2009;81:171–6. Gilbert RE, Topkey TA, Cubitt WD, Ades AE, Masters J, Peckham CS. Prevalence of toxoplasma IgG among pregnant women in west London according to country of birth and ethnics group. BMJ 1993;306:185. Cook AJC, Gilbert RE, Buffolano W, Zufferey J, Peterson E, Jenum A, et al. Sources of toxoplasma infection in pregnant women: European multicentre case-control study. BMJ 2000;321:142–7. Jones JL, Kruszon-Moran D, Wilson M. Toxoplasma gondii infection in the United States, 1999–2000. Emerg Infect Dis 2003;9:1371–4. Raude J. La place de la viande dans le mode`le alimentaire franc¸ais. Cah Nutr Diet 2008;43:1.
www.sciencedirect.com Revue d’E´pide´miologie et de Sante´ Publique 61 (2013) 311–317
Original article
Prevalence of toxoplasmosis in France in 1998: Is there a difference between men and women? At what age do children become infected? La pre´valence de la toxoplasmose en France en 1998 est-elle diffe´rente entre les hommes et les femmes ? A` quel aˆge s’infectent les enfants ? H. Bellali a,b,*, H. Pelloux c, I. Villena d, H. Fricker-Hidalgo c, Y. Le Strat a, V. Goulet a,* a
De´partement des maladies infectieuses, institut de veille Sanitaire, 12, rue Val-d’Osne, 94415 Saint-Maurice cedex, France b Profet – Programme de formation a` l’e´pide´miologie de terrain, 94415 Saint-Maurice cedex, France c Laboratoire de parasitologie-mycologie, universite´ Joseph-Fourier, centre hospitalier universitaire de Grenoble, 38000 Grenoble, France d Centre national de re´fe´rence de la toxoplasmose, universite´ de Reims – Champagne-Ardenne, SFR Cap-Sante´, centre hospitalier universitaire de Reims, 51100 Reims, France Received 10 January 2012; accepted 15 November 2012
Abstract Background. – The only national seroprevalence data currently available on toxoplasmosis in France are from the national perinatal surveys of pregnant women conducted in 1995 and 2003. These surveys are national, exhaustive and cross-sectional studies of all pregnant women who give birth in France during one specified week. These cross-sectional studies, conducted among women of childbearing age (defined as 18 to 45 years), showed a positive correlation between seroprevalence and age, with a significant regional disparity. This study was performed in order to compare the prevalence of toxoplasmosis antibodies in men and women in the 18–45 age group, to confirm regional variations and to estimate the seroprevalence of toxoplasmosis in France for different age groups, particularly among children and among adults aged over 45 years. Methods. – Serum samples from 2060 subjects were available from a national serum bank that was established in 1997 as part of a European study on vaccine preventable diseases. The sera were tested for IgG antibodies in 2008–2009, by ELISA test, at the laboratory of parasitologymycology, CHU Grenoble. Results. – The seroprevalence for the population aged 1–64 years was 55.4%. Seroprevalence did not vary between the sexes, except among those aged over 45 years, where it was higher in men than in women. Toxoplasmosis seroprevalence varied significantly by regions for all ages. It increased with age and we noted a stronger increase in prevalence in adolescents (10–20 years) than in other age groups. Conclusion. – This study showed that children have limited exposure to Toxoplasma gondii and that seroprevalence in men and women does not differ for the population aged 45 years and under. This study confirms the geographical disparity in prevalence in France that has been found in other studies in women of childbearing age. This disparity cannot be explained by different laboratory techniques, because sera were tested in a single laboratory. The study also raises the possibility of extrapolating seroprevalences from ENP to the general population and thus estimating the seroprevalence in the French population. # 2013 Published by Elsevier Masson SAS. Keywords: Ttoxoplasmosis; Antibodies; Prevalence; France
Re´sume´ Position du proble`me. – Les seules donne´es nationales de la se´ropre´valence de la toxoplasmose disponibles en France a` ce jour proviennent des enqueˆtes nationales pe´rinatales re´alise´es en 1995 et 2003 chez la femme enceinte. Ces enqueˆtes nationales, qui sont des e´tudes transversales exhaustives re´alise´es pour toutes les naissances en France durant une semaine, ont montre´ une corre´lation positive de la se´ropre´valence de la toxoplasmose avec l’aˆge et une disparite´ re´gionale importante. Les objectifs de ce travail e´taient de comparer la pre´valence entre les hommes et les femmes en aˆge de procre´er (18–45 ans), de ve´rifier les variations entre les re´gions et d’estimer la se´ropre´valence de la toxoplasmose en France selon l’aˆge, notamment chez les enfants et les adultes aˆge´s de plus de 45 ans.
* Corresponding author. E-mail address: [email protected], [email protected] (H. Bellali), [email protected] (V. Goulet). 0398-7620/$ – see front matter # 2013 Published by Elsevier Masson SAS. http://dx.doi.org/10.1016/j.respe.2012.11.005
312
H. Bellali et al. / Revue d’E´pide´miologie et de Sante´ Publique 61 (2013) 311–317
Me´thodes. – Un e´chantillon de se´rum concernant 2060 sujets provenait d’une se´rothe`que nationale mise en place dans le cadre d’une e´tude se´roe´pide´miologique europe´enne re´alise´e en 1997 pour e´tudier les maladies a` pre´vention vaccinale. La recherche des anticorps antitoxoplasmose type IgG a e´te´ faite graˆce au test Elisa en 2008–2009 au laboratoire de Parasitologie-Mycologie, CHU de Grenoble. Re´sultats. – La se´ropre´valence de la toxoplasmose chez la population aˆge´e de un a` 64 ans e´tait de 55,4 %. Elle n’e´tait pas diffe´rente entre les hommes et les femmes a` l’exception de ceux qui sont aˆge´s de plus de 45 ans ou` elle e´tait plus e´leve´e chez les hommes. Elle variait significativement selon les re´gions et elle augmentait avec l’aˆge, avec une augmentation importante a` l’adolescence. Conclusion. – Cette e´tude montre que les enfants sont peu expose´s a` la toxoplasmose et que jusqu’a` l’aˆge de 45 ans la se´ropre´valence des hommes n’est pas significativement diffe´rente de celle des femmes. Cette e´tude confirme que la disparite´ ge´ographique de la se´ropre´valence de la toxoplasmose en France de´ja` de´crite ne peut pas eˆtre explique´e par des diffe´rences de techniques entre laboratoires car tous les se´rums ont e´te´ teste´s en un seul lieu. Elle permet de conside´rer que les se´ropre´valences de l’Enqueˆte nationale pe´rinatale peuvent eˆtre extrapole´es aux hommes et donnent des estimations de la pre´valence de la toxoplasmose dans la population ge´ne´rale. # 2013 Publie´ par Elsevier Masson SAS. Mots cle´s : Toxoplasmose ; Anticorps ; Pre´valence ; France
1. Introduction Toxoplasmosis is a zoonotic disease caused by the protozoan parasite Toxoplasma gondii [1]. This infection is prevalent in humans and animals worldwide [2]. Felids are the key animal species in the life cycle of this parasite because they are the hosts that can excrete the environmentally-resistant stage, the oocyst [3]. The prevalence of T. gondii varies greatly in populations throughout the world and seems to be affected by food-production practices, water treatment, climate, topography, soil and cat feces-related hygiene, occupational and nonoccupational soil exposure, and culinary practices. The infection is more prevalent in some regions in Europe and parts of South America than in Asia, the United States or Australia [4–6]. In humans, infection is usually either asymptomatic or the cause of mild flu-like symptoms. However, toxoplasmosis can be life threatening in immunocompromised individuals. Moreover, if acquired during pregnancy, toxoplasmosis can lead to a wide variety of manifestations in the fetus and infant, including spontaneous abortion, stillbirth, or a live infant with classic signs of congenital toxoplasmosis such as hydrocephalus or microcephalus, cerebral calcifications and retinochoroiditis [2]. A prevention program for congenital toxoplasmosis was implemented in France in 1978. Systematic serological screening of all pregnant women has been undertaken since 1985. Monthly monitoring was implemented in 1992. It involves follow-ups of toxoplasmosis serology for all non-immune pregnant women. These women receive primary prevention advice on food hygiene, food preparation and avoiding eating certain types of foods [7]. Diagnosis of maternal seroconversion during pregnancy is necessary to prevent transmission to the fetus, by means of specific treatment and also to permit close monitoring of at-risk fetuses and neonates. A surveillance program for congenital toxoplasmosis has been operating in France since 2006, and is coordinated by the National Reference Center for toxoplasmosis [8]. The only current national seroprevalence data available in France are from the national perinatal surveys (ENP) conducted in 1995 and 2003 [9,10]. ENP surveys are national, exhaustive and cross-sectional studies of all pregnant women who give birth in France during one specified week. These studies,
conducted among women of childbearing age (defined as 18 to 45 years), showed a positive correlation between seroprevalence and age, with a significant regional disparity. The main objectives of this study were to compare the prevalence of toxoplasmosis antibodies between men and women in the 18–45 age group in order to establish whether the results from national perinatal surveys (ENP) could be extrapolated to the general population and thus estimate the prevalence in the French population, and to check geographic variations. The second objective was to estimate toxoplasmosis seroprevalence according to age, mainly among children and adults over the age of 45.
2. Material and methods In 1997, a cross-sectional study was carried out on sera collected from a sample of the French population within the framework of the European sero-epidemiology network (ESEN), which coordinates and harmonizes serological surveillance of immunity to vaccine preventable diseases. National banks of serum specimens have been collected in six European countries (Denmark, England, France, Germany, Italy and the Netherlands) to provide about 3500 specimens for each country, stratified by age, with males and females equally represented [11]. In France, specimens were collected at 39 volunteer centers (30 private biomedical laboratories, three hospital biomedical laboratories and six hospital departments), between January 1998 and July 1999. Subjects were recruited by biomedical laboratories from people who came to the laboratory for a blood sample for any reason. After consent, additional tubes of blood were collected for the purposes of the study. During the sampling, the biologist collected information from each individual on sex, age, administrative district of residence, country of birth, health insurance coverage, and month and year of sampling. For adults, education level and occupational category were recorded, and for children (< 18 years), the occupational category of the head of household. Each laboratory had a quota of persons by age and sex to sample. Subjects with conditions that could affect their immunity were excluded from the ESEN survey. These conditions included being immunocompromised; having received a blood
H. Bellali et al. / Revue d’E´pide´miologie et de Sante´ Publique 61 (2013) 311–317
transfusion or having been treated with gamma globulin or other blood products in the previous 6 months; and having hepatic failure, renal failure, nephrotic syndrome, hematologic diseases or cancers, malnutrition or autoimmune diseases. The final sample underrepresented people living in rural areas and over-represented people in the highest sociooccupational categories. The conserved sera were tested blind for toxoplasmosis antibodies in 2008–2009 (Fig. 1). Serological tests for toxoplasmosis were performed by the laboratory of parasitology-mycology, CHU Grenoble, using ELISA Vidas IgG-II (bioMe´rieux, Marcy l’Etoile, France) and AxSYM IgG (Abbott Laboratories, Wiesbaden, Germany) [12,13]. When the results obtained using the two ELISA techniques were discordant, an IgG in-house immunofluorescence [14] was performed to make a decision on the presence or absence of antibodies against T. gondii. Data were analyzed using STATA version 9.2. Age was grouped into six classes for univariate analysis: 1–9 years, 10– 19 years, 20–29 years, 30–39 years, 40–49 years and 50–64 years. For multivariate analysis, age was introduced as a continuous variable modeled by a fractional polynomial. To take into account the interaction between age and sex, three categories (< 18 years, 18–45 years and 46–64 years) were used to present stratified results of multivariate analysis. The study did not include French Overseas districts. France is divided into eight major areas of territory study and planning (Zones d’e´tudes et d’ame´nagement du territoire [ZEAT]), using the territorial division created in 1967 by the national institute of statistics and economic studies (Institut national de la statistique et des e´tudes e´conomiques [INSEE]). Each ZEAT contains between one and six administrative regions. This division provides areas that are fairly homogeneous, with a sufficient number of subjects to test for differences in seroprevalence. Areas of residence were coded into one of three categories: rural area, which was defined as a town with a population under 2000; urban area with fewer than 200 000 inhabitants; and urban area with 200 000 or more inhabitants, including Greater Paris.
[(Fig._1)TD$IG]
Seroepidemiological survey 1998: 2937 sera collected 688 not available (insufficient amount)
Residual sera: 2249 sera Excluding those born outside France, n=185 And those <1 year, n=4
Study sample: 2060 subjects Fig. 1. Source and number of sera used in toxoplasmosis study, France 1998.
313
In 1998, the prevalence in France for the population aged between 1 and 64 years was estimated by applying the observed prevalence in our study to the French population aged 1–64 years (data INSEE 1998), stratified by age, region and, for those aged 46 years and over, by sex. For each ZEAT, age-specific prevalence rates (by 5 year age groups) and age- and sexspecific rates (for those aged over 46 years) were applied to ZEAT population and summed. The total number of seropositives was then divided by the 1998 total French population aged 1–64 years, to calculate the overall seroprevalence. To compare the results of this study with those of the National Perinatal Survey (ENP) 1995, prevalence was estimated for women aged between 18 and 45 years, by standardizing for age and region of the survey sample 1995 (ENP). We studied the association of toxoplasmosis with the following variables: age, sex, occupational category, ZEAT region and area of residence. Comparison of seroprevalence by age, sex, region of residence, occupation and residential area, was performed first by univariate analysis, then by a multivariate backward logistic regression model and testing possible interactions between all the independent variables. All variables studied were included in the model for multivariate analysis. This choice was justified by the low number of factors studied. Age was introduced as a continuous variable and was modeled using fractional polynomial, giving a relationship between toxoplasmosis seroprevalence and age [15]. 3. Results A total of 2060 sera were tested for IgG toxoplasma antibodies. Data were missing for between four and 96 individuals, depending on the variable. The average age was 24 years (range: 1–64 years, median: 20 years), and men accounted for 49% of subjects in the sample. In 1998, the prevalence of toxoplasmosis in France for the population aged between 1 and 64 years was 55.4% (95% CI: 53.4–57.4%). Prevalence for women aged between 18 and 45 years, after standardizing for age and region of the sample population 1995 of the ENP, was estimated at 55.7% (95% CI: 54.7,56.7%). Table 1 shows population and toxoplasmosis seroprevalence by sociodemographic characteristics. Toxoplasmosis seroprevalence increased with age. It was 4% in children under 10 years and 88% in people aged 50 and over (P < 0.001). Between the ages of 10–19 years, seroprevalence was 7.5 times greater than that observed in children aged less than 10 years. The prevalence of toxoplasmosis varied significantly between rural (31%) and urban (47%) areas (P < 0.001). It did not differ between men and women (45.3% and 41.6% respectively, P = 0.081). Multivariate analysis results are presented separately by age category because of a significant (P = 0.023) interaction between age as a continuous variable and sex (Table 2). Toxoplasmosis seroprevalence varied significantly by area whatever the age. A low prevalence was observed in the
314
H. Bellali et al. / Revue d’E´pide´miologie et de Sante´ Publique 61 (2013) 311–317
Table 1 Study population distribution and seroprevalence of toxoplasmosis by the factors studied, cross-sectional study, France 1998. na
Seroprevalence (n = 896)
Crude odds ratio (OR)
438 560 406 191 185 278
3.9 30 49.5 61.8 79.5 88.1
Reference 10.1 23.1 38.1 91.1 174.9
Sex Male Female
1006 1048
45.3 41.6
1.2 Reference
Occupational category Craftsman/farmer Manual/non-manual workers Professional/executive Other/inactive Student Retired
87 1025 348 231 185 115
16.1 47.1 21.3 47.2 49.7 79.1
Reference 4.6 1.4 4.6 5.1 19.8
2.6–8.4 0.7–2.6 2.4–9.0 2.6–10.1 7.8–50.3
270 341 108 222 277 233 383 225
29.6 43.4 47.2 32.4 39.7 48.1 54 51.1
Reference 1.8 2.1 1.1 1.6 2.2 2.8 2.5
1.3–2.6 1.3–3.4 0.8–1.7 1.1–2.2 1.5–3.2 2.0–3.9 1.7–3.6
212 430 1322
31.6 35.3 47.6
Reference 1.2 1.9
0.8–1.7 1.4–2.7
Age (years) 1–9 10–19 20–29 30–39 40–49 50–64
ZEAT regions Centre-East Paris Basin North East West South-West Greater Paris Mediterranean region Residential area Rural (< 2000 inhabitants) Urban (< 200 000 inhabitants) Urban ( 200 000 inhabitants and Greater Paris) a
95% CI
P
< 0.001 5.9–17.2 12.7–41.9 18.4–78.7 35.3–234.9 56.4–541.8 0.9–1.4
0.081
< 0.001
< 0.001
< 0.001
Data were missing for between 1 (ZEAT regions) and 96 (Residential area) individuals, depending on the variable.
Centre-East (30.7%), a high prevalence in Greater Paris (57.5%) and the South-West (52.3%) and medium prevalence (35–50%) in other areas (East, West, Paris Basin, North and the Mediterranean Region) (Fig. 2). In children (< 18 years) and in adults (18–45 years), toxoplasmosis prevalence increased with increasing age, but did not vary between sexes. Over 45 years, a significantly higher prevalence was observed in men (OR = 2, P = 0.033) than in women. In multivariate analysis, once the interaction between sex and age was taking into account, neither occupational category nor living in a rural or urban district was associated with toxoplasmosis. In order to represent the prevalence age profile, we calculated estimated prevalence for each year of age and among men and women. We represented the obtained curve with observed data (Fig. 3). 4. Discussion In 1998, the prevalence of toxoplasmosis in France in the population aged between 1 and 64 years was estimated at 54%. Toxoplasmosis seroprevalence increased with age, with a more
marked rise during adolescence. Prevalence did not differ between men and women between the ages of 18 to 45 years but was higher in men over 45 years than in women in the same age group. These sera were collected from individuals attending medical laboratories, so were not a random sample of the population designed for a French seroepidemiological survey for vaccine preventable diseases. This sample underestimated people living in rural areas and over-represented people in the highest socio-occupational categories. These two factors were not, however, associated with toxoplasmosis prevalence in multivariate analysis. The distribution of these two factors was similar between men women. Seroprevalence among women aged 18–45 years was estimated at 55.7%, which is similar to the seroprevalence observed in the 1995 ENP study (54.3%) [9]. The highest prevalences were observed in Greater Paris. The South-West and Centre-East regions of France had the lowest prevalence. National perinatal surveys conducted in 1995 [9] and 2003 [10] showed the same regions of high and low prevalence. The results of toxoplasmosis serology collected in national perinatal studies were produced by routine laboratories, which used
H. Bellali et al. / Revue d’E´pide´miologie et de Sante´ Publique 61 (2013) 311–317
315
Table 2 Risk factors of toxoplasmosis seroprevalence in France, 1998: results of multivariate analysis by age category. Age category 1–17 years (n = 869) Variables
OR
95% CI
Age (in years) 1–4 5–9 10–14 15–17 18–24 25–29 30–34 35–39 40–45 46–49 50–54 55–59 60–64
1 3.1 9.6 23.6 – – – – – – – – –
2.3–3.9 4.5–14.6 6.2–41.0 – – – – – – – – –
Sex Male Female
0.98 1
ZEAT regions Centre-East Paris Basin North East West South–West Greater Paris Mediterranean region
1 2.5 0.9 1.3 1.4 3.6 3.9 3.8
18–45 years (n = 838)
46–64 years (n = 350)
P-value
OR
95% CI
P-value
OR
95% CI
P-value
< 0.001 0.170 < 0.001 < 0.001 –
– – –
– – – –
–
– – – – – – – – – 1 0.98 0.95 0.93
– – – – – – – – –
–
0.7–1.3 0.4–1.5 0.1–1.8
0.875 0.088 0.187 0.881
1.1–4.0
0.033
2.3–25.0 1.1–15.9 0.7–6.1 1.2–8.8 2.2–140.4 2.8–29.2 2.2–30.6
< 0.001 0.001 0.04 0.2 0.022 0.007 < 0.001 0.002
–
1 1.4 1.9 2.5 3.2 – – – –
0.6–1.5
0.9
1.3 1
1.0–6.1 0.2–4.7 0.5–3.6 0.5–3.8 1.4–9.0 1.6–9.5 1.5–9.5
< 0.001 0.037 0.902 0.626 0.468 0.006 0.002 0.005
1 2.8 2.17 1.9 2.2 3.7 5.1 2.93
1.3–1.6 1.5–2.3 1.7–3.2 1.9–4.5 – – – –
< 0.001 0.080 0.002 0.034 < 0.001 –
0.9–1.7
0.08
2 1
1.7–5.0 1.2–4.6 1.1–3.4 1.3–3.8 2.1–6.6 3.1–9.0 1.6–5.1
< 0.001 < 0.001 0.017 0.033 0.005 < 0.001 < 0.001 < 0.001
1 7.6 4.1 2.1 3.2 17.4 9 8.1
(Fig._2)TD$IG][OR: adjusted odds ratio.
Fig. 2. Seroprevalence of toxoplasmosis by area, standardized for age and sex, general population, France, 1998.
–
[(Fig._3)TD$IG] 316
H. Bellali et al. / Revue d’E´pide´miologie et de Sante´ Publique 61 (2013) 311–317 1 0,9
Seroprevalence
0,8 0,7 0,6 0,5 0,4 0,3
Observed prevalence
0,2
Estimated prevalence among women
Estimated prevalence
Estimated prevalence among men
0,1 0 0
10
20
30
40
50
60
70
Age (years)
Fig. 3. Estimated prevalence of toxoplasmosis by age and sex, France 1998. h i h i Equation for men: prevalenceðageÞ ¼ exp 7:79 ððage þ 1Þ=10Þ1=2 0:619 = 1 þ exp 7:79 ððage þ 1Þ=10Þ1=2 0:619 Equation for women: prevalenceðageÞ ¼ expð2:01 ½Inððage þ 1Þ=10Þ 0:918Þ=ð1 þ expð2:01 ½Inððage þ 1ÞÞ=10 0:918ÞÞ.
different techniques and reagents with potentially variable sensitivities. If testing habits did vary between regions, this would have an impact on the prevalence found in the study. In our study, regional disparities cannot be explained by different laboratory techniques because the sera were tested in the same laboratory. This local variation is most likely to be related to culinary practices and the ability of oocysts to survive in different climates. A positive correlation between regional consumption of mutton and toxoplasmosis prevalence has been shown by Ancelle et al. in ENP 1995 [9] and Berger et al. in ENP 2003 [10]. Halos et al. [16] showed that districts which had high seroprevalence in sheep were the same districts that had high seroprevalence in women of childbearing age in ENP surveys. This observation could be explained by an intense circulation of the parasite in these areas and may be due to specific environmental conditions (such as density of felids, climate characteristics and dietary habits). Interestingly, the geographical disparities are observed from childhood with a non-significant increase with age. This study is important because to our knowledge it is the first in France to estimate and compare the prevalence of toxoplasmosis in males and females, and in children and in the elderly. Seroprevalence was very low in children under 10 years old (4%). This low prevalence was also reported in a study conducted recently in a rural region of France (2.5% estimated seroprevalence in 7-year-olds) [17]. In light of these results, the transmission of toxoplasmosis from soil contaminated by cat feces does not seem to be the major route of contamination in France, since contact with the ground starts in early childhood (3–5 years). This is the first French study to provide prevalence data for men, because most previous studies have investigated toxoplasmosis in pregnant women. In general, prevalence does not vary greatly between the sexes [18–24]. In our study, seroprevalence did not vary between the sexes, except in men and women over 45 years old, where it was higher in men. It is known that the prevalence of T. gondii antibodies increases with age, as has been found by several studies [5,9,10,17,18,25–30]. This reflects the increased probability of having been in contact with the parasite through one of the transmission routes as people get older.
We noted an important increase in prevalence during adolescence. The same finding has been observed in many other studies [19,25,27–29]. Fernandes et al. [25] estimated the average age of acquisition of toxoplasmosis at 11 years, through mathematical modeling based on a seroepidemiological study in a Brazilian population between 1984 and 2006. Taylor et al. [19] found a nearly linear increase in seroprevalence between 10 and 14 years of age in urban and rural schoolchildren aged 4 to 18 years in Dublin, Wicklow and Kildare (Ireland). In Doha (Qatar), Abu-Madi [27] observed a prevalence that increased from 6% in the 2–10 year age group to 16% in the 11–20 year age group. SalahiMoghaddam [28] reported an increase in toxoplasmosis prevalence of almost 30% from the 5–9 year age group to the 10–14 year age group. In a community-based survey of human toxoplasmosis in rural Amazonia [29], the log-probit model fitted to age-prevalence data estimated the median age of seroconversion at 14 years. These observations may be explained by a change in eating habits of children of this age, with a higher consumption of meat including mutton or lamb, known to be a main source of transmission of Toxoplasma worldwide and especially in Europe [31,32]. Moreover, studies of consumption of meat in France (INCA 1999 and CCAF 2003) have shown a steady increase in the average daily consumption of meat with age and a tendency in young people (15–24 years) to prefer eating red meat (beef and lamb) [33]. The study reported here allowed us to estimate changes in seroprevalence by age between 1 and 64 years. A similar age profile of estimated probability of seropositivity was described by Fromont et al. [17] in a toxoplasma seroprevalence study in the French rural population in 2004 and also by Ferreira et al. [29] in a communitybased survey of human toxoplasmosis in rural Amazonia, Brazil. Prevalence did not differ between men and women between the ages of 18 to 45 years. These results show that the estimated seroprevalence surveys in ENP can be extrapolated to the general population using the estimated curve of seroprevalence by age.
H. Bellali et al. / Revue d’E´pide´miologie et de Sante´ Publique 61 (2013) 311–317
Disclosure of interest The authors declare that they have no conflicts of interest concerning this article.
[17]
[18]
References [1] Ryan KJ, Ray CG, Sherris JC. Toxoplasmosis. In: Sherris medical microbiology: an introduction to infectious diseases4th ed., United States: McGraw Hill; 2004. p. 723–7. [2] Montoya JG, Liesenfeld O. Toxoplasmosis. Lancet 2004;363:1965–76. [3] Hutchison WM, Dunachie JF, Siim JC. Cats as a source of toxoplasmosis. Proc R Soc Med 1972;65:1001. [4] McQuillan G, Kruszon-Moran D, Kottiri BJ, Curtin LR, Lucas JW, Kington RS. Racial and ethnic differences in the seroprevalence of 6 infectious diseases in the United States: data from NHANES III, 1988– 1994. Am J Public Health 2004;94:1952–8. [5] Jones JL, Kruszon-Moran D, Wilson M, McQuillan G, Navin T, McAuley JB. Toxoplasmosis in the United States: seroprevalence and risk factors. Am J Epidemiol 2001;154:357–65. [6] Jones JL, Dargelas V, Roberts J, Press C, Remington JS, Montoya JG. Risk factors for Toxoplasma gondii infection in the United States. Clin Infect Dis 2009;49:878–84. [7] Carme B, Tirard-Fleury V. La toxoplasmose chez la femme enceinte en France : se´ropre´valence, taux de se´roconversion et niveau de connaissance des mesures pre´ventives. Tendances 1965–1995. Med Mal Infect 1996;26:431–6. [8] King L, Villena I, Ancelle T, Wallon M, Thulliez P, et al. La toxoplasmose conge´nitale : mise en place d’un dispositif de surveillance en France. Bull Epidemiol Hebdomadaire 2008;14-15:122–4. [9] Ancelle T, Goulet V, Tirard-Fleury V, Baril L, Mazaubrun C, Thulliez P, et al. La toxoplasmose chez la femme enceinte en France en 1995. Re´sultats d’une enqueˆte nationale pe´rinatale. Bull Epidemiol Hebdomadaire 1996;51:227–9. [10] Berger F, Goulet V, Le strat Y, Desenclos JC. Toxoplasmosis among pregnant women in France: risk factors and change of prevalence between 1995 and 2003. Rev Epidemiol Sante Publique 2009;57:241–8. [11] Osborne K, Weinberg J, Miller E. The European sero-epidemiology network. Eurosurveillance 1997;2:29–31. [12] Fricker-Hidalgo H, Saddoux C, Suchel-Jambon AS, Romand S, Foussadier A, Pelloux H, et al. New VIDAS assay for toxoplasma-specific IgG avidity: evaluation on 603 sera. Diagn Microbiol Infect Dis 2006;56:167–72. [13] Gay-Andrieu F, Fricker-Hidalgo H, Sickinger E, Espern A, BernierPinchart MP, Braun HB, et al. Comparative evaluation of the architect toxo IgG, IgM avidity assays for anti-toxoplasma antibodies detection in pregnant women sera. Diagn Microbiol Infect Dis 2009;65:279–87. [14] Roberts A, Hedman K, Luyasu V, Zufferey J, Bessie`res MH, Blatz RM, et al. Multicenter evaluation of strategies for serodiagnosis of primary infection with Toxoplasma gondii. Eur J Microbiol Infect Dis 2001;20:467–74. [15] Royston P, Ambler G, Sauerbrei W. The use of fractional polynomials to model continuous risk variables in epidemiology. Int J Epidemiol 1999;28:964–74. [16] Halos L, The´bault A, Aubert D, Thomas M, Perret C, Geers R, et al. An innovative survey underlining the significant level of contamination by
[19]
[20]
[21]
[22]
[23]
[24]
[25]
[26]
[27]
[28]
[29]
[30]
[31]
[32] [33]
317
Toxoplasma gondii of ovine meat consumed in France. Int J Parasitol 2010;40:193–200. Fromont EG, Riche B, Rabilloud M. Toxoplasma seroprevalence in a rural population in France: detection of a household effect. BMC Infect Dis 2009;9:76–82. 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. Taylor MRH, Lennon b, Holland CV, Cafferkey M. Community study of toxoplasma antibodies in urban and rural schoolchildren aged 4 to 18 years. Arch Dis Child 1997;77:406–9. Alvarado-Esquivel C, Mercado-Suarez MF, Rodriguez-Briones A, FelladTaures L, Alaya-Alaya JO, Nevarez-Piedra LJ, et al. Sero-epidemiology of infection with Toxoplasma gondii in healthy blood donors of Durango, Mexico. BMC Infect Dis 2007;7:75–82. Kook J, Lee HJ, Kim B, Yun ChK, Guk SM, Seo Min. et al. Toxoplasma gondii antibody titers in sera of children admitted to the Seoul National University Children’s Hospital. Korean J Parasitol 1999;37:27–32. Lee YH, Noh HJ, Hwang SK, Shin DW. Seroepidemiological study of Toxoplasma gondii infection in the rural area Okcheon-gun, Korea. Korean J Parasitol 2000;38:251–6. Shin DW, Cha DY, Hua QJ, Cha GH, Lee YH. Seroprevalence of Toxoplasma gondii infection and characteristics of seropositive patients in general hospitals in Daejeon, Korea. Korean J Parasitol 2009;47: 125–30. Yang HJ, Park YK, Hong SC, Bae GM, Lee SH. Seroprevalence of toxoplasmosis in the residents of Cheju Island, Korea. Korean J Parasitol 2000;38:91–3. Fernandes GCVR, Azevedo RS, Amaku M, Yu ALF, Massad E. Seroepidemiology of toxoplasma infection in a metrpolitan region of Brazil. Epidemiol Infect 2009;137:1809–15. Ertug S, Okyay P, Turkmen M, Yuksel H. Seroprevalence and risk factors for toxoplamsa infection among pregnant women in Aydin province, Turkey. BMC Public Health 2005;5:66–71. Abu-Madi M, Al-Molawi N, Behnke JM. Seroprevalence and epidemiological correlates of Toxoplasma gondii infections among patients referred for hospital-based serological testing in Doha, Qatar. Parasites Vectors 2008;1:39–47. Salahi-Moghaddam A, Hafizi A. A serological study on Toxoplasma gondii infection among people in south of Tahran, Iran. Korean J Parasitol 2009;47:61–3. Ferreira MU, Hiramoto RM, Aureliano DP, Da Silva-Nunes M, Da Silva NS, Malafonte RS, et al. A community based survey of human toxoplasmosis in rural Amazonia: seroprevalence, seroconversion rate and associated risk factors. Am J Trop Med Hyg 2009;81:171–6. Gilbert RE, Topkey TA, Cubitt WD, Ades AE, Masters J, Peckham CS. Prevalence of toxoplasma IgG among pregnant women in west London according to country of birth and ethnics group. BMJ 1993;306:185. Cook AJC, Gilbert RE, Buffolano W, Zufferey J, Peterson E, Jenum A, et al. Sources of toxoplasma infection in pregnant women: European multicentre case-control study. BMJ 2000;321:142–7. Jones JL, Kruszon-Moran D, Wilson M. Toxoplasma gondii infection in the United States, 1999–2000. Emerg Infect Dis 2003;9:1371–4. Raude J. La place de la viande dans le mode`le alimentaire franc¸ais. Cah Nutr Diet 2008;43:1.