Even today, the diagnosis of fetal toxoplasmosis infection presents a serious challenge to obstetricians, because diagnosis implies invading the fetal environment to obtain fluid. The decision to perform amniocentesis is based on the analysis of maternal serology, which is not always simple to interpret due to the different methodologies used in clinical pathological examinations and serial repetition of serological tests in different laboratories when in doubt [1].

There are many studies that have sought a method to diagnose fetal infection as early as possible. Diagnosis has evolved over time from amniocentesis and inoculation of the amniotic fluid in mice to cordocentesis, with non-specific and specific assessment of the fetal blood, and more recently to the use of the polymerase chain reaction (PCR) method applied to fetal fluids, especially amniotic fluid [2, 3].

This evolution has led to the fulfilment of the true role of the specialist in maternal-fetal medicine: not only to assess the natural course of the disease through ultrasound, but also to take action to preserve the fetus.

Given the aspects presented above, and the fact that an infected fetus is a potentially high-risk patient and deserves differentiated postnatal care, the authors sought to assess the PCR to diagnose fetal toxoplasmosis infection in the amniotic fluid.

A retrospective longitudinal study was conducted by analysing the medical records of pregnant women in the Department of Obstetrics as well as the newborns in the nursery at the Neonatal Paediatrics Division of the Federal University of São Paulo (UNIFESP). This study was approved by the UNIFESP Research Ethics Committee.

Inclusion criteria were as follows: pregnant women of any age, parity and gestational age with specific positive serology (IgM) for toxoplasmosis, either reactive or with values above the cutoff level, with indications for amniocentesis and PCR of amniotic fluid, along with the possibility of post-natal follow-up. Exclusion criteria were as follows: withdrawal of the couple during interviews or examinations prior to amniocentesis, multiple gestation, missing the post-natal appointment, fetal death unrelated to toxoplasmosis, termination of pregnancy without formal indication, and patients using spiramycin prior to amniocentesis.

Data about serology (IgM), transfontanellar ultrasound, and examination of the fundus were taken from the medical records of newborns whose mothers had undergone amniocentesis to perform PCR on the amniotic fluid, confirming or dismissing congenital infection (whether manifesting or not) during hospital-alization.

Interviews were conducted with all the pregnant women with specific IgM indicating recent and/or current maternal toxoplasmosis in their charts. After the interview, the details and risks of the procedure were explained, and ultrasound was performed to confirm gestational age of >14 weeks and absence of alterations in the amniotic membrane and placental site. Next, amniocentesis was performed. The material was immediately taken to the laboratory, and the patient was directed to rest for 24 hours and begin treatment with spiramycin (1,000 mg every eight hours), continuing this treatment until returning for the results. Subjects with negative PCR continued spiramycin until the end of pregnancy, and underwent serial ultrasound every three weeks. Positive cases initiated triple therapy with sulfadiazine, pyrimethamine, and folinic acid, alternating every three weeks with spiramycin. The ultrasound was also repeated at three-week intervals. The newborn was considered to have had positive contact with the etiologic agent if changes were seen in the transfontanellar ultrasound, fundus or specific serology (IgM) for acute toxoplasmosis.

Statistical analysis was performed with descriptive determination of sensitivity, specificity, and positive and negative predictive values. Statistical significance was defined by Fischer’s exact test, with error fixed at < 0.05 to reject the null hypothesis. The Epi Info statistics program was used to calculate the tests.

The mean gestational age at amniocentesis was 20.61 ± 4.41 weeks. The mean gestational age at amniocentesis was 20.61 weeks, with no statistically significant difference between the results of the punctures performed before or after the 20^th week. Of the 84 newborns who underwent transfontanellar ultrasound, changes were only observed in four cases (4.8%).

The positive PCR rate for the etiologic agent of toxoplasmosis was 17.9% (15 patients) and the rate of newborns in contact with this agent was 16.7% (14 infants). Of these 14 infants, five manifested the disease, while nine of them only had contact with T. gondii with no signs of toxoplasmosis until the time of hospital discharge.

Of the five newborns with the disease, three were born to women who had a negative pre-natal PCR. There was a rate of 83.3% false positive maternal serology. The PCR had a sensitivity of 78.6%, specificity of 94.3%, positive predictive value of 73.3%, and negative predictive value of 95.6% (Tables 1 and 2).

Toxoplasmosis is caused by T. gondii and becomes more important when the affected host is a pregnant woman. Infected pregnant women are asymptomatic in most cases, but the disease can cause irreversible harm to the fetus such as microphthalmia, cerebral calcifications, microcephaly, hydrocephalus, mental retardation, hepatosplenomegaly, chorioretinitis, prematurity, growth restriction, and even foetal death [4].

The first choice for maternal treatment is spiramycin; in addition to this agent, materno-fetal treatment uses sulfadiazine, folinic acid, and pyrimethamine, thereby avoiding vertical transmission and fetal sequelae in around 50% of cases [5].

In this study, the patients with positive serology for toxoplasmosis (IgM) were submitted to amniocentesis at different gestational ages, with an average of 20.61 weeks. Among the PCR tests performed before the 20^th week, 16.6% (8/48) were positive. Among those performed in the 20^th week or later, 19.4% (7/36) were positive, showing that there was no significant statistical difference between the two groups (p = 0.74). Thus, it became clear that the gestational age at the time of the puncture did not influence the results, which led the authors to question if vertical transmission truly increased with gestational age.

Treating the fetuses with positive PCR using the alternate scheme was seen to modify neonatal outcome, preventing greater disease involvement in the treated infants compared to the untreated infants. When the authors evaluated the four cases of newborns who had no contact with the causative agent, but presented positive PCR (false positive) in the prenatal examination, they questioned whether the false positives were the result of laboratory contamination of the amniotic fluid sample or whether maternal and fetal treatment had been effective.

The fact that only 16.7% of newborns had evidence of contact with the causative agent, regardless of whether the disease was expressed or not, and that 83.3% of newborns with no evidence of contact with T. gondii were born to seropositive mothers, i.e. false positives, led the authors to consider two possibilities: maternal serological false-positives and adequate maternal treatment preventing vertical transmission.

It can be inferred that not all fetuses who had contact with the agent presented the disease due to interference from effective maternal-fetal treatment, reducing the number of sick newborns by 66.8%.

Of the study population, when considering the newborn results (i.e. contagion with the aetiologic agent or not), the authors obtained a sensitivity result of 78.6% and a specificity result of 94.3%, with a positive predictive value of 73.3% and a negative predictive value of 95.6% for the PCR test. The results of this study were similar to those in the literature, showing PCR to be an effective method for prenatal diagnosis of fetal contact with the etiological agent, regardless of whether the newborn manifests toxoplasmosis or not [6].

In toxoplasmosis, time is a determining factor in fetal prognosis. When the clinical presentation does not permit correct interpretation of maternal serology in inconclusive situations, assistance can be obtained as quickly as possible through this method instead of repeating new serologic testing, which causes more confusion than clarification and results in a loss of precious time in maternal-fetal treatment.