Congenital toxoplasmosis is a rare (NSC 2001) but potentially severe parasitic infection that can lead to intrauterine death or stillbirth, malformation, mental retardation, deafness and blindness of the infected infant (Montoya 2004). It is caused by Toxoplasma gondii (T. gondii). T. gondii is one of the most common infectious pathogenic animal parasites of man, belonging to the phylum apicomplexa group (Montoya 2004). Other members of this phylum include known human pathogens such as Plasmodium (malaria) and Cryptosporidium. It is acquired by ingesting oocysts excreted by cats, contaminated soil or water, or by eating the undercooked meat of infected animals, which contain tissue cysts (Gilbert 2002; Montoya 2004). Most cases of toxoplasmosis infection are asymptomatic and self‐limited except for congenital infection and immunocompromised patients (Montoya 2004); hence many cases remain undiagnosed. The incubation period of acquired infection is estimated to be within the range of four and 21 days (seven days on average) (Rorman 2006). Serological surveys demonstrate that world‐wide exposure to T. gondii is high (30% in US and 50% to 80% in Europe) (Rorman 2006). The susceptibility of pregnant women (that is the rate of seronegative pregnant women) to toxoplasmosis varies between countries. It is up to 90% in northern Europe, where T. gondii is not so common (Cook 2000; Gilbert 2002). When infection does occur during pregnancy, T. gondii can be transmitted from the mother to the fetus (vertical transmission) and can lead to congenital toxoplasmosis. Multiple factors are associated with the occurrence of congenital toxoplasmosis infection, including route of transmission, climate, cultural behaviour, eating habits and hygenic standards (Rorman 2006). The probability of transmission of the parasite to the fetus varies according to the gestational age and the risk is greater during the third trimester (from 5% at 12 weeks to 80% just before delivery) (Gilbert 2002). Conversely, the severity of the condition, that is the risk of the fetus to develop major clinical signs, decreases with increasing gestational age (from 60% at 12 weeks to 5% just before delivery) (Gilbert 2002). Clinical features include hydrocephalus (excessive accumulation of cerebrospinal fluid within the cranium), microcephaly (abnormal smallness of the head, usually associated with mental retardation), deafness, cerebral calcifications, seizures and psychomotor retardation. Signs of a systemic infection may also be present at birth, including fever, rash, and enlargement of liver and spleen. Fetal infection can cause inflammatory lesions of the retina and choroids that can lead to visual impairment. Moreover, it can cause lesions of the brain leading to mental damage; more rarely the infection can cause the death of the fetus or of the newborn (Gilbert 2002; Montoya 2004; NSC 2001). Severe damage in infancy occurs in 5% of congenital toxoplasmosis cases while intracranial or ocular lesions are observed in 20% to 30% of cases by three years of age (Gilbert 2002). Although there is no consensus on the most appropriate screening or treatment for congenital toxoplasmosis, three possible approaches have been proposed: prenatal screening, neonatal screening and primary prevention (Gilbert 2002).

Prenatal screening
Prenatal screening (secondary prevention), offered in some European countries, for example France, Switzerland, Germany, Austria and Italy where the incidence of T. gondii maternal infection is more frequent, is based on the timely detection of the mother's infection by a serum test for toxoplasma IgG and IgM (NSC 2001). If the first prenatal test shows signs of recent infection or a seroconversion is detected during pregnancy, a confirmatory test is required before starting treatment with spiramycin or pyrimethamine‐sulfadoxine or both (Foulon 1999a; Gilbert 2002; Montoya 2004). Diagnosis of fetal infection is performed by amniocentesis, which is known to be associated with 1% risk of miscarriage (Alfirevic 2003) and testing of the amniotic fluid for the detection of the parasite or, most recently, of toxoplasma DNA by polymerase chain reaction (PCR) technique. Congenital toxoplasmosis can also be diagnosed by cordocentesis, that consists in drawing fetal blood from the umbilical cord, and the detection of the parasite or specific immunoglobulin (IgM and IgA) in the fetal blood, but the risk of complications due to the procedure is higher (Bader 1997; Foulon 1999b; Gilbert 2002). If fetal infection is confirmed, the parents can decide either to terminate the pregnancy or to opt for drug treatment. Prenatal screening, although advocated by some as essential for reducing congenital toxoplasmosis (Boyer 2005), has several limitations: false‐positive toxoplasma IgM results are common, false‐positive toxoplasma IgG are less common but also possible (Gilbert 2002; Montoya 2004); moreover, the rate of false‐positive test results can increase notably in settings where local prevalence of the infection is lower; there can be organizational problems or problems of acceptability due to the need to repeat the serum test every four to six weeks in seronegative women (Bader 1997); there is no evidence on the efficacy of antenatal treatment in reducing transmission to the fetus nor of improving neonatal outcomes or reducing functional impairment in later childhood (Montoya 2004; Peyron 1999); there are problems concerning the accuracy of the diagnostic test for fetal infection, particularly the lack of a standardized technique for PCR (Chabbert 2004; Foulon 1999b; Thalib 2005). Finally, this strategy causes additional fetal losses, that are healthy fetuses lost due to amniocentesis and to elective terminations of pregnancy: it has been estimated that the number of additional losses necessary to prevent one additional case of toxoplasmosis can be as high as 18.5 in case of universal screening in a setting with a low incidence of T. gondii maternal infection like the USA (Bader 1997).

Neonatal screening
Neonatal screening (tertiary prevention), adopted in Poland, Denmark and some areas of the USA (NSC 2001), consists of the diagnosis of newborn infection through detection of toxoplasma specific IgM on Guthrie card blood spots. In fact, up to 90% of infected infants are asymptomatic at birth and will show clinical symptoms only in later life (Stegmann 2002). Current guidelines suggest that infected infants should receive treatment with pyrimethamine and sulfadiazine for up to one year regardless of symptoms (Gilbert 2002). Even if this strategy is technically feasible and less costly than prenatal screening, there is no evidence that treating the infected children has any effect (Gilbert 2002; Lebech 1999). Moreover, this approach is ineffective on irreversible damages already present at birth. Considering such limitations, neonatal screening should be adopted only in places where other options are not available; the implications of such a policy should be fully discussed with the parents of the newborn tested.

Primary prevention
Primary prevention can involve the whole population, by educating the general public and filtering water, and veterinary public health intervention (such as labeling to indicate toxoplasma‐free meat and improved farm hygiene to reduce animal infection): this will reduce the protozoan circulation and could be an option but up to now there is not enough research to determine the feasibility and efficacy of this approach (Gilbert 2002; NSC 2001). Another possibility is primary prevention based on prenatal education of pregnant women or women of reproductive age to avoid toxoplasmosis in pregnancy (Gilbert 2002). In fact, sources and risk factors for contracting toxoplasmosis are well known (Cook 2000) and can be avoided by adopting simple behavioral measures like not eating raw or insufficiently cooked meat, washing hands thoroughly after handling raw meat and after gardening, avoiding contact with cats' faeces (directly or indirectly through the soil, or possibly contaminated raw vegetables or fruits) (Cook 2000; Gilbert 2002). Primary prevention based on prenatal education, if proven to be effective, could be a good strategy to reduce congenital toxoplasmosis, since it will not imply any of the problems linked to secondary and tertiary prevention strategies discussed above.

Readers may wish to refer to the following Cochrane systematic reviews for further information about toxoplasmosis in pregnancy: 'Timing and type of prenatal treatment for congenital toxoplasmosis' (Thiébaut 2003) and 'Treatments for toxoplasmosis in pregnancy' (Peyron 1999).