Section of Epidemiology, National Institute of Public Health, PO Box 4404 Nydalen, 0403 Oslo, Norway.
Testing for antibodies against Toxoplasma gondii in pregnancy is routinely offered in some countries. However, no randomized controlled trials of the effect of treatment have been performed, and the question of whether testing in pregnancy should be encouraged rests on evidence from observational studies. Two such studies are presented in this issue. The first investigated the effect of timing and type of treatment on the risk of vertical transmission.1 The authors hypothesize that the lack of effect in their study is explained by rapid transmission to the fetus after maternal infection, implying that the initiation of treatment comes too late.
The other study estimates the impact of timing and type of treatment on the risk of intracranial and ocular lesions in congenitally infected children at 3 years of age.2 Even after controlling for the length of gestation at maternal infection, they found a paradoxical effect of treatment. A long interval from seroconversion to treatment lowered the risk of intracranial lesions, whereas the opposite would be expected. The interval seemed to have no impact on the risk of ocular lesions and the type of treatment did not seem to influence the risk of congenital damage. The authors state that their results provide no evidence that prenatal screening is beneficial.
The French Studies in Context
The finding of no treatment effect on the transmission rate is in agreement with a recent European multicentre study including 144 infected women,3 and with systematic literature reviews.4,5 However, estimates of transmission rates vary widely. In the present study,1 the proportion of infected children was 28%, while it was 19% in a large population-based Danish study.6 In the multicentre study,3 the transmission rate varied from 30% to 73% between centres, with an overall value of 44%. This variability raises issues of inter-study differences in selection of cases or differences in laboratory methods.
The finding in the second study2 that treatment has no beneficial effect on sequellae is in contrast to the results from the multicentre study,3 where a significant, protective effect of treatment is reported. The studies differ in their analytical approach. The study by Gras et al.2 includes only outcomes in children where transmission has occurred (i.e. the children who are at direct risk of sequellae), while the multicentre study3 includes children of all seroconverting women in the analysis (the outcome is a combination of risk of transmission and risk of sequellae given transmission). If the latter study3 had limited the analysis to children under direct risk, it appears from inspection of the reported data that the protective effect would have been smaller and insignificant, although this cannot be directly judged since the distribution of gestational age at infection (the main confounder) is not presented by treatment group. A major problem with all the observational studies is the selection of patients to treatment, and given treatment, the selection to short or long time intervals before treatment starts.
Implications for Testing in Pregnancy
The observational studies discussed above do not provide convincing evidence for beneficial effects of prenatal treatment of toxoplasmosis. Treatment in pregnancy should, in our opinion, be regarded as experimental, and only be performed as part of carefully conducted randomized trials.
Even if there were a beneficial effect of treatment in pregnancy, there may not be an overall positive effect of screening. The effect of screening depends on the magnitude of the health problem, the estimated treatment effect, but also on the compliance to the screening programme and the treatment.5 Also, possible side effects of treatment must also be considered, as discussed by Gras et al. on the basis of their findings.2 Pyrimethamine is a folic acid antagonist. The use of folic acid antagonists in pregnancy has been associated with increased risk of neural tube defects.7 Termination of pregnancy with a healthy fetus and complications to invasive prenatal diagnosis are other important potential side effects. In addition to the economic costs and the use of limited health resources, which traditionally have been in focus when debating screening initiation, the psychological aspects have come increasingly into focus.8 Initial false positive diagnoses are common when screening for disorders of low prevalence. Both false and verified positive diagnoses may cause anxiety in the mother and her family throughout the pregnancy and reduce the positive expectations for the new child, even though the risk of a severely diseased child is low.
On this background, should one commence health district-randomized controlled trials on the effect of introducing a screening programme of testing in pregnancy? In today's situation, with little evidence of any beneficial effect of treatment, our opinion is no. First, one should document that treatment is effective.
Future Research
We suggest that the effect of prenatal treatment should be tested out using an ordinary double-blind placebo-controlled clinical trial with randomization on an individual basis.
Additionally, better estimates of the burden of disease should be made, including population statistics. How many children in a certain population will suffer from the consequences of congenital toxoplasmosis? The observational studies give little detail as to the degree of disability and illness experienced by these children. Such data are also needed for evaluation of public health actions.
We also suggest that controlled community trials could be performed to estimate the effect of primary prevention directed against established risk factors.9 Thus, one could randomize health care districts to have intervention or no intervention, where the intervention might consist of detailed advice to women in early pregnancy to modify behaviour with respect to consumption of raw or undercooked meat and unwashed vegetables, and behaviour with respect to contact with cat faeces.
References
1
Gilbert RE, Gras L, Wallon M, Peyron F, Ades AE, Dunn DT. Effect of prenatal treatment on mother to child transmission of Toxoplasma gondii: retrospective cohort study of 544 mother-child pairs in Lyon, France. Int J Epidemiol 2001;30:130308.
2
Gras L, Gilbert RE, Ades AE, Dunn DT. Effect of prenatal treatment on the risk of intracranial and ocular lesions in children with congenital toxoplasmosis. Int J Epidemiol 2001;30:130913.
3 Foulon W, Villena I, Stray-Pedersen B et al. Treatment of toxoplasmosis during pregnancy: a multicenter study of impact on fetal transmission and children's sequelae at one year of age. Am J Obstet Gynecol 1999; 180:41015.[ISI][Medline]
4
Wallon M, Liou C, Garner P, Peyron F. Congenital toxoplasmosis: systematic review of evidence of efficacy of treatment in pregnancy. Br Med J 1999;318:151114.
5 Eskild A, Oxman A, Magnus P, Bjørndal A, Bakketeig LS. Screening for toxoplasmosis in pregnancy: what is the evidence of reducing a health problem? J Med Screening 1996;3:18894.[Medline]
6 Lebech M, Andersen O, Christensen NC et al. Feasibility of neonatal screening for toxoplasma infection in the absence of prenatal treatment. Lancet 1999;353:183437.[ISI][Medline]
7
Hernandez-Diaz S, Werler MM, Walker AM, Mitchell AA. Neural tube defects in relation to use of folic acid antagonists during pregnancy. Am J Epidemiol 2001;153:96168.
8
Stewart-Brown S, Farmer A. Screening could seriously damage your health. Br Med J 1997;314:53334.
9 Kapperud G, Jenum PA, Stray-Pedersen B, Melby K, Eskild A, Eng J. Risk factors for toxoplasma infection in pregnancy. Results from a prospective study in Norway. Am J Epidemiol 1996;144:40512.[Abstract]