Toxoplasmosis is a parasitic infection that is usually asymptomatic or mildly symptomatic, except for infections in immunocompromised patients, including congenital patients.1,2 In congenital toxoplasmosis, the severity of infection depends mainly on gestational age at the time of maternal infection with resulting vertical transmission and possibly on the parasite genotype, the host genetic predisposition, and the parasite load in amniotic fluid (AF) samples.2–7
Conventional polymerase chain reaction (PCR) has been used in laboratory diagnosis to detect the presence of Toxoplasma gondii DNA in AF samples and less frequently on fetal or neonatal blood. In some centers, quantitative PCR has replaced conventional PCR as a result of better sensitivity and predictive values.6,8,9 Costa et al3 analyzed AF samples by quantitative PCR and found a trend associating high parasite load with fetal cranial ultrasound abnormalities and early infection, whereas Romand et al4 inferred that parasite load greater than 100 parasites/mL was associated with severe infections when gestational age at maternal infection was less than 20 weeks.
The aim of this study was to retrospectively correlate T gondii load detected by quantitative PCR using stored DNA from AF samples, gestational age at the time of maternal infection, and the presence of anti–T gondii immunoglobulin (Ig) M at birth to the signs and severity of congenital infections (clinical outcome) at birth and at 12 months follow-up.
MATERIALS AND METHODS
The present study has a prospective and a retrospective approach. The prospective part of the research refers to the quantitative PCR performed on stored DNA extracted from AF samples. This prospective part was approved by the Commission for Analysis of Research Projects of the Clinical Board of the Clinical Hospital, School of Medicine, University of Sao Paulo (protocol number 0347/11) and was performed in Sao Paulo, Brazil, during a 3-year period (2012–2014). The retrospective part of the study had been previously approved by the same ethical commission (protocol number 0326/07), performed conventional PCR on DNA from AF samples, and constituted a biorepository of AF samples and DNA. The study has also analyzed data from medical records of pregnant women (prenatal care and delivery), neonates at birth, and infants who were monitored up until the end of the first year of life.
Pregnant women received prenatal care in public referral hospitals of the city of Sao Paulo and only those who attended at least five consultations were recruited. As such, third-trimester maternal infection is excluded from this study.
Routine screening for maternal anti–T gondii IgG and IgM was performed at the first prenatal visit. Positive results on automated commercial kits were confirmed by an in-house indirect immunofluorescence test.10 Thereafter, 149 women were subjected to IgG avidity tests through Vidas Toxo IgG Avidity to estimate gestational age at the time of maternal seroconversion to toxoplasmosis. Interpretation of IgG avidity results followed the manufacturer's recommendation: low avidity less than 0.2; gray zone 0.2 or greater and less than 0.3; high avidity 0.3 or greater. Low avidity results indicate that toxoplasmosis has been recently acquired, within the previous 12 weeks; high avidity indicates a previous infection, acquired more than 12 weeks ago, and results that fall in the gray zone are not conclusive. Considering that one third of results fell into the gray zone, the avidity test was performed a second time to allow more reliable dating. A total of 122 women repeated the test 4 weeks later, and, after the second testing, gestational age at the time of maternal infection was estimated. The avidity results were analyzed together with the last menstrual period and fetal biometric parameters determined by ultrasonography. Instead of using a number referring to the weeks of gestation ±1–2 weeks, we adopted the estimated week without the uncertainty interval to be able to perform the statistical analyses. This procedure has already been used in other studies.4,11
Amniocentesis is part of the clinical protocol used to investigate pregnant women who seroconvert to toxoplasmosis. The procedure was performed under ultrasound guidance by fetal medicine specialists during the second trimester (16–25 weeks of gestation). At the time of amniocentesis, most pregnant women had already received spiramycin (3 g/d) for 4 weeks (in only four women, amniocentesis was performed before this period of time) following the Brazilian Health Ministry guidelines,12 which recommends treatment as soon as the serologic results are known, before amniocentesis, up to 1 week before the expected date of delivery. The treatment protocol was the same in the three participant hospitals. The obstetric teams received the conventional PCR results within 72 hours after the sample collection. Although PCR is the gold standard of laboratory diagnosis, spiramycin was not discontinued if qualitative PCR was negative because the negative predictive value of PCR is not 100%.13 On the other hand, pregnant women with positive PCR results attended more consultations and laboratory and ultrasound examinations until labor.
Pregnant women were informed of the amniocentesis risks and the amplification procedures; they agreed with all laboratory testing that would be performed on them and on their children, and they gave written informed consent.
As part of a diagnostic routine procedure offered to all of the children of mothers diagnosed with primary toxoplasmosis acquired during pregnancy, after birth, children return for health checkups at the end of the first week of life, at the end of the second week, and thereafter once a month for pediatric evaluations. Neurologic and ophthalmologic consultations took place every 3 months, as well as laboratory, radiology, and other procedures. In the pediatric evaluation, general health conditions of children were recorded as well as their nutritional status and the acquisition of abilities according to that expected for their age. In addition, measurements of weight and height and of the cephalic, thoracic, and abdominal perimeters were performed. In the neurologic evaluation, transfontanelle cranial ultrasonography and eventually computed tomography (CT) scans were carried out and the Bayley Scales for Infant Development-II14 instrument was applied to assess the infant neuropsychomotor development for the early detection of deficits. The eye fundus examination was performed for the detection of chorioretinitis.
The confirmation of congenital toxoplasmosis took place at birth through the presence of anti–T gondii IgM, associated or not with signs suggestive of toxoplasmosis, transfontanelle cranial ultrasonography, CT scan results, and eye fundus abnormalities. In IgM-negative neonates at birth, the appearance of IgM after birth or the persistence of IgG anti–T gondii associated or not with signs suggestive of infection by 12 months of age was also considered as evidence of congenital infection.6,13
The follow-up of 12-month-old infants was performed by three specialists, a neonatologist, an ophthalmologist, and a neurologist who were aware of the diagnosis of toxoplasmosis made by serologic and conventional PCR results. Beginning at birth, and continuing every 3 months until the end of the first year of life, other laboratory tests such as the complete blood count, liver enzymes (transaminases, alkaline phosphatase, gamma glutamyl transferase, albumin) were performed in addition to antitoxoplasmosis serology, cranial transfontanelle ultrasonography, CT scans, and eye fundus. Computed tomography scan was not performed in all patients but was mandatory when there was suspicion of central nervous system involvement. All of the children were treated for 12 months, as continuously as possible, using the same protocol: pyrimethamine (1 mg/kg/d), sulfadiazine (100 mg/kg/d), and folinic acid (10–15 mg, three times a week). Prednisone at a dose of 1 mg/kg/d was prescribed when there was neurologic, severe ophthalmologic involvement, or both.15,16
Neonatal and infant signs and symptoms were defined by physical examination, imaging, and laboratory analysis. Chorioretinitis varied from single and small lesions with little inflammatory process to extensive injuries involving the macular region. Transfontanelle cranial ultrasound abnormalities included cerebral calcifications, ventriculomegaly, and hydrocephaly. Other signs of infection were fever or hypothermia, hepatosplenomegaly, jaundice with a predominance of direct bilirubin, skin rash, myocarditis, and respiratory distress as well as laboratory abnormalities including complete blood count, liver enzymes, and total IgM. Severely compromised fetuses were those presenting with cranial ultrasound abnormalities and severely compromised neonates were those with chorioretinitis involving the macular region associated or not with cranial ultrasound abnormalities.
DNA extracted from AF samples were tested by T gondii conventional PCR17 (performed during pregnancy) and quantitative PCR (performed with stored DNA samples, after the end of the follow-up). T gondii quantitative PCR was performed as previously described.8 Briefly, 100 ng of DNA were used in amplifications containing 0.4 micromolar of B22 and B23 primers from the parasite B1 gene. AF samples from pregnant women without acute toxoplasmosis were used as negative controls as well as sterile water (nontemplate sample). The 115-base pair quantitative PCR product was cloned into the QIAGEN-PCR Cloningplus kit vector and subsequently used as the positive control. A calibration curve with seven serial dilutions of the positive control showed a detection limit of one third of a single parasite. Considering that 2 mL of the DNA samples containing 100 ng of DNA were tested, quantitative PCR results were expressed in parasites/mL by multiplying the parasite load found by 500 and then dividing this result by 35, because each parasite contains 35 copies of the B1 gene.18
Inclusion criteria of the study were: 1) infants whose mothers had positive anti–T gondii serology with amniocentesis done in the second trimester and positive qualitative PCR results; 2) gestational age at the time of maternal infection determined by the last menstrual period, fetal biometric parameters in ultrasonography, and maternal IgG avidity index; and 3) infants with complete clinical (pediatric, neurologic, ophthalmologic) and laboratory (imaging and serologic examinations) 12-month follow-up. Exclusion criteria included: 1) AF sample containing blood or meconium or of insufficient volume (less than 3 mL) and 2) incomplete infant follow-up.
Demographic and laboratory data were used to build a databank using Microsoft Excel 2010 software. The Spearman correlation coefficient was used to determine whether the variables gestational age at the time of maternal infection, presence of neonatal positive anti–T gondii IgM at birth, and the parasite load level in the second-trimester AF samples were associated with signs of infection in the neonates at birth and in the infants up to 12 months of age. This analysis was performed using the Graph Pad Prism 6.0.
A logistic regression analysis was performed considering as the dependent variable (primary outcome) the presence or absence of signs of congenital toxoplasmosis from birth until the end of the 12-month follow-up. This variable was divided into two categories: 1 for presence and 0 for absence of signs of infection. The independent variables were gestational age at the time of maternal infection considering ranks with 2-week increments, parasite load levels determined by quantitative PCR in AF samples arbitrarily divided into ranks with increments of 20 parasites/mL, and the presence of positive anti–T gondii IgM at birth. This last variable was divided into two categories: 1 for IgM positive and 0 for IgM negative. The logistic regression analysis was performed using SPSS Statistics 22.
The parasite load levels were also distributed according to percentile and symptomatology (symptomatic or asymptomatic and severe infection or not).
Three receiver operator characteristic curves were plotted and their corresponding area under the curves (AUCs) were calculated to test the value of gestational age at maternal infection, parasite load levels in AF samples, and the two parameters combined regarding the clinical outcome (presence or absence of signs of infection) of fetuses and infants with congenital toxoplasmosis.
The study recruited 122 women who attended prenatal care at the three reference hospitals and agreed to participate. Demographic and social characteristics of the 122 pregnant women and the infants were described elsewhere.7 Briefly, 85 (69%) women were between 20 and 34 years old; 56 (45%) were white, 47 (39%) had mixed ethnicity (mixture of black and white), and 19 (16%) were black. Sixty-four of 122 (52%) declared to have attended school for 5–8 years. Sixty-five of 122 (53%) were primigravida. Fifty-five (45%) women reported household contact with cats or dogs. The consumption of meat was reported by 38 (31%), 49 (40%), and 35 (29%) women once, twice, and three times a week, respectively. Eighteen (15%) reported consumption of raw or undercooked meat during pregnancy. Family income in 74 (60%) households was up to one minimum wage (the equivalent of 300 U.S. dollars monthly). Piped, treated water was available in the home in 80% and sewerage systems in 61%.
The neonatal gestational age was estimated at birth by the Capurro method for full-term neonates19 or the Alexander method for preterm neonates.20 Birth weights ranged from less than 2,500 g in 14 (11%) to 58 (48%) between 2,500 and 3,000 g, and 50 (41%) were greater than 3,000 g. Of the 122 neonates, 100 (82%) were full term (37–42 weeks of gestation) and 22 (18%) preterm (30–36 weeks of gestation). Among the full term, 43 of 100 were small for gestational age (43%), and there were 8 of 22 (36.3%) small-for-gestational-age neonates in the preterm group.
Overall, 86 of 122 (70.5%) infants did not show signs of congenital toxoplasmosis by the end of the 12-month follow-up. Of the 36 (29.5%) with evidence of congenital infection, 26 of 36 (72.2%) had abnormal fetal ultrasound findings. These included hepatosplenomegaly (n=21), brain calcifications (n=4), hydrocephalus (n=2), and severe fetal growth restriction less than the fifth percentile (n=2). There were no cases of microcephaly. Some fetuses had more than one abnormal finding.
At birth, 107 of 122 (87.7%) had neonatal anti–T gondii IgM antibodies present, and 86 of 122 (70.5%) were asymptomatic. Of these, 72 of 86 (83.7%) had neonatal anti–T gondii IgM antibodies present, and among the 36 who ever developed symptoms, 35 of 36 (97.2%) had neonatal anti–T gondii IgM antibodies present (Table 1). All of the 36 participants who ever developed signs or symptoms of congenital toxoplasmosis did so at birth. The 15 of 122 (12.3%) patients that had neonatal anti–T gondii IgM absent at birth were in the asymptomatic group. They had the diagnosis of congenital toxoplasmosis confirmed during the follow-up by means of the onset of IgM anti–T gondii in one infant at 6 months or the persistence of IgG at the end of the 12-month follow-up, providing evidence that the infant began to produce antibodies raised to T gondii.
Neonatal findings among the 36 symptomatic neonates included hepatosplenomegaly in 21 (58.3%), respiratory distress in eight (22.2%), jaundice in six (16.6%), hypothermia in six (16.6%), fever in five (13.8%), and skin rash in one (2.8%). No neonate developed myocarditis. Liver enzymes were increased in 15 (41.7%) infants, anemia was found in 20 (55.5%), and leukocytosis in 18 (50%). Among the 36 symptomatic neonates, six (16.7%, 4.9% of the total) had severe infections (central nervous system involvement associated or not with chorioretinitis involving the macula); two of these six infants presented the Sabin's triad of symptoms (chorioretinitis, hydrocephalus, and brain calcifications).21 Among the six severely compromised neonates, all had leukopenia, four had thrombocytopenia, and three had high total IgM.
Infection occurred in 18 of 122 mothers (14.7%) in the first trimester (up to 12 weeks of gestation) and 104 of 122 (85.3%) during the second trimester (13–25 weeks of gestation). Among the 36 infants with signs of infection, 18 were in the group infected during the first trimester and 18 belonged to the group infected during the second trimester. All of the 18 fetuses infected during the first trimester had symptoms, and among them, there were six severely compromised.
The distribution of fetuses and infants according to gestational age at the time of maternal infection (X-axis) considering the signs and severity of infections are displayed with respect to the parasite load level expressed in the logarithmic scale (Y-axis; Fig. 1). Figure 2 shows the week of maternal infection for each of the 36 symptomatic neonates and how they are distributed throughout the first two trimesters of pregnancy.
Among the 122 participants, parasite load levels were highly variable, with a median of 35 parasites/mL (range 2–30,473 parasites/mL). Twenty participants (16%) had parasite loads greater than 100 parasites/mL, of whom 19 (95%) had signs of infection (including the six patients with severe infection) (Table 1). Of the 36 infants with signs of infection, 19 (52.8%) had AF parasite loads greater than 100 parasites/mL; only 1 of the 86 infants without signs of infection by 12 months of age (1%) had greater than 100 parasites/mL (147 parasites/mL).
All of the six severely compromised neonates were infected at early gestational ages (four during the first and two in the early second trimester) and had parasite loads greater than 100 parasites/mL. Three of the highest parasite load levels were found among these six neonates; two of them (30,473 and 6,191 parasites/mL) belonged to infants with the Sabin's triad of symptoms since birth, also associated with hypothermia, anemia, and thrombocytopenia. The third highest parasite load level (1,647 parasites/mL) belonged to a neonate with hepatosplenomegaly, pronounced jaundice, and abnormal cranial ultrasonogram and CT scan (brain calcifications) without chorioretinitis. The other three severely compromised fetuses had parasite load levels of 197, 218, and 428 parasites/mL. One of them had brain calcifications, and the other two had chorioretinitis involving the macula associated with pronounced and prolonged jaundice, hepatosplenomegaly, and leukocytosis. All signs of infection were detected at birth.
There were two infants with unexpected outcomes. The first had a high parasite load (16,167 parasites/mL), corresponding to the second highest of the study, and was infected at an estimated gestational age of 25 weeks. This infant had intense and prolonged jaundice but had normal neurologic development, cranial ultrasonograms, CT scans, and eye fundus at 12 months. The second patient with an unexpected outcome had a parasite load of 147 parasites/mL, gestational age of infection was estimated at 14 weeks, and there were no signs of infection up to 12 months.
Table 2 shows the distribution of the 122 parasite load levels according to percentile (5th, 25th, 50th, 75th, 90th, 95th, and 100th) as well as the presence and severity of symptomatology. Up to the 75th percentile (parasite load 57.75 parasites/mL), all of the patients totaling 62 were asymptomatic. In the 75th percentile, there were 30 participants (20 asymptomatic and 10 symptomatic). In the 90th percentile (parasite load 161.4 parasites/mL), there were 18 participants (four asymptomatic and 14 symptomatic); in the 95th percentile (parasite load 214.9 parasites/mL), there were six participants, all of them symptomatic and one severely compromised. Lastly, in the 100th percentile (parasite load 30,473.0 parasites/mL), there were six participants, all of them symptomatic and five severely compromised. Interestingly, the 36 symptomatic participants appeared from the 75th percentile (parasite load 57.75 parasites/mL) and the six severely compromised participants from the 95th percentile (parasite load 214.90 parasites/mL).
The logistic regression analysis (Table 3) has shown that signs of infection are correlated with gestational age at maternal infection (adjusted odds ratio [OR] 0.47, P<.001, 95% CI 0.31–0.73), meaning that for every 2-week increment in gestational age at maternal infection, the odds of symptomatic infection is approximately half. Regarding the parasite load, for every increment of 20 parasites/mL, the odds doubled for the presence of signs of infection (adjusted OR 2.04, P=.006, 95% CI 1.23–3.37). There was no association between positive IgM results at birth and the presence of signs of infection (OR 6.81, P=.069, 95% CI 0.86–53.9).
A negative correlation coefficient was observed between gestational age at the time of maternal infection and the parasite load (rs −0.780, P<.001, 95% CI −0.843 to −0.696, Spearman correlation coefficient) as well as between gestational age and signs of infection (rs −0.664, P<.001, 95% CI −0.755 to −0.547). No correlation was found between gestational age and positive IgM results (rs −0.136, P=.135, 95% CI −0.311 to 0.048).
The three receiver operator characteristic curves and their corresponding AUC are shown in Figure 3. Briefly, when gestational age at maternal infection was analyzed with respect to symptomatology of infants, AUC was 0.918 (95% CI 0.855–0.960). Regarding the parasite load levels and symptomatology, AUC was 0.959 (95% CI 0.908–0.987) and when both parameters were combined, AUC was 0.969 (95% CI 0.920–0.992).
The laboratory diagnosis of congenital toxoplasmosis during prenatal care is based on maternal serology detecting anti–T gondii IgG and IgM antibodies in addition to possible changes in fetal ultrasonography, but the gold standard for the laboratory diagnosis since the end of the 20th century is the amplification of T gondii DNA in AF samples, mainly by PCR.13 Initially, conventional PCR was performed generating positive or negative results, and this qualitative result is enough to establish whether fetuses are infected. During the past two decades, technologic advances have gradually replaced conventional PCR with quantitative amplifications. This presents numerous advantages over conventional techniques because it is faster, more sensitive, less subject to contamination resulting from carryover, and quantitative PCR avoids the manipulator exposure to substances that intersperse the DNA and are carcinogenic such as ethidium bromide.22 Nevertheless, it is important to point out that the main advantage of quantitative PCR is undoubtedly its ability to quantify the pathogens present in the biologic sample in the context of infectious diseases. This feature is especially important in infections for which treatment is available, opening up possibilities for monitoring the efficacy of therapy. Although we are not proposing that pregnant women should undergo more than one amniocentesis, the first evaluation of the parasite load in AF samples can be useful to establish the fetal, neonatal, and postnatal risk of symptomatic and severe congenital toxoplasmosis.
Quantitative amplifications have replaced conventional in toxoplasmosis studies,22 but most reports still use the results qualitatively, that is, they restrict the analysis to the positivity or negativity of samples. Only Costa et al3 and Romand et al4 have analyzed quantitative PCR results performed on DNA extracted from AF samples of pregnant women who acquired toxoplasmosis during pregnancy in a quantitative way. Costa et al3 studied 87 AF samples and found seven parasite load results greater than 1,000 parasites/mL, of which four belonged to severely infected fetuses. According to the neonatal cranial ultrasound data, there were six severely compromised infants (6.9%), four had ventriculomegaly, were infected during the first trimester, and had parasite load levels greater than 1,000 parasites/mL. The authors concluded that there was a trend associating high parasite loads with fetal cranial ultrasound abnormalities and early infection.
Romand et al4 investigated 88 congenital infections and found 12 neonates (13.6%) with ventriculomegaly or cerebral calcifications, a fourfold increment in comparison with Costa et al3 and this study. They showed a significant negative linear regression between gestational age at maternal infection and T gondii loads in AF samples. After adjusting for time at maternal seroconversion by a multivariate analysis, parasite loads greater than 100 parasites/mL were associated with severe infections, but only when gestational ages at maternal infection were less than 20 weeks.
To further investigate the advantages of quantitative PCR in the laboratory diagnosis of congenital toxoplasmosis, the present study has prospectively tested stored DNA from AF samples by quantitative PCR. These samples had been previously tested by a conventional PCR. At the time of the conventional PCR testing, neonates and infants were clinical and laboratory monitored up until the end of the first year of life. In this way, we were able to determine whether signs of infection detected from birth to the end of the 1-year follow-up were associated with gestational age at maternal infection, the presence of anti–T gondii IgM in neonates, and the parasite load determined on AF samples by quantitative PCR.
Amniocentesis was performed from 16 weeks of gestation, although there is a recommendation to perform the procedure from 18 weeks of gestation, at least 4 weeks after the probable date of maternal seroconversion, to avoid false-negative results. The amount of DNA recovered after AF samples extraction is very low, and the analysis of samples from gestation ages earlier than 18 weeks generates even lower amounts of DNA, sometimes rendering the amplifications unfeasible.13
The logistic regression analysis (Table 3) has demonstrated that, not only are signs of infection highly correlated with gestational age at vertical transmission, but parasite load levels are also correlated with signs of infection, whereas IgM results at birth are not. Moreover, the Spearman correlation coefficient showed a highly significant negative correlation between gestational age at the time of maternal infection and the parasite load, as expected, and the same level of significance was attained when gestational age and signs of infection were correlated. Again, no significance was found between gestational age and IgM results. The logistic regression results as well as the correlation coefficient results have shown that the parasite load levels are associated with the clinical outcome of neonates and infants with congenital toxoplasmosis, and this association does not depend on gestational age at the time of maternal infection.
In the present study, it was possible to show the validity of replacing a conventional qualitative PCR with a quantitative amplification. When the parasite load levels were distributed according to percentile and symptomatology, it was possible to observe that the parasite load value and its percentile may help the clinical management of infected pregnant women and their concepts because symptomatic fetuses and infants were only detected from the 75th percentile, corresponding to a parasite load of 57.75 parasites/mL, and all the severely compromised patients were concentrated from the 95th percentile, corresponding to a parasite load of 214.90 parasites/mL.
Romand et al4 associated the severity of congenital toxoplasmosis with parasite loads greater than 100 parasites/mL, but only when the maternal infection occurred before 20 weeks of gestation, differing from the results of the present study that were determined independently of gestational age at maternal infection. If we consider the analysis performed in this study according to parasite load percentile, the value suggested by Romand et al4 of 100 parasites/mL falls between the 75th percentile from which we observed the onset of symptomatology in neonates and infants and the 95th percentile from which all the severe infections were found. It is important to note that Romand et al4 have also used B1 gene primers, probably the same primer pair of this study, although the quantitative PCR did not apply the same protocol and laboratory facilities.
Another way of exploring the results of the present study was to estimate the risk for the fetus, neonate, and infant to present symptomatic infection by means of the parasite load based on load rankings, which has demonstrated a doubled OR for each 20-parasite/mL increment. This type of analysis can be helpful in clinical practice.
The ability to estimate at what gestational age the infection occurred is important to make decisions on the treatment and surveillance of the suspected patients. When the fetus is infected, the risk for clinical symptoms at birth or during the first years of life decreases with increasing gestational ages at maternal infections.23 This effect of gestational age on the symptomatic or asymptomatic nature of congenital toxoplasmosis has been persistently demonstrated over the past decades.24–26 On the contrary, only a few studies have attempted to use quantitative amplification techniques (quantitative PCR) to predict the outcome of congenital toxoplasmosis. Romand et al4 determined that maternal infections acquired before 20 weeks of gestation with a parasite load in AF of more than 100 parasites/mL are associated with a very high risk (OR 15.4, 95% CI 2.45–98) of severe fetal infection. In the present study, when the three AUC of receiver operator characteristic curves were analyzed, it was noteworthy that parasite load levels were associated with symptomatic infections (AUC 0.959, 95% CI 0.908–0.987) as much as gestational age (AUC 0.918, 95% CI 0.855–0.960) and both parameters combined (AUC 0.969, 95% CI 0.920–0.992). In case the attending physician can use both parameters (parasite load and gestational age at maternal infection), the maximum association will be attained. Once again, this type of analysis can be helpful in clinical practice.
If the parasite load levels in AF samples are going to help medical decisions, the same primer pair from the same target gene performed with the same quantitative PCR protocol has to be used to generate results that can be interpreted according to the established distribution of parasite loads in percentile. Each prenatal care center can standardize its own quantitative PCR technique either using the same target gene of this study (B1 gene) described by Burg et al18 or the repetitive 529-base pair gene27 that seems to be currently preferred. Alternatively, prenatal care centers can use a commercial quantitative PCR kit such as the one analyzed in a recent multicenter study,22 the Toxoplasma ELITe MGB designed with REP-529-base pair primers.27 This quantitative PCR kit was compared with three in-house quantitative PCR assays of three laboratories belonging to the French National Reference Center for Toxoplasmosis network. Calibrated parasite suspensions with concentrations ranging from 0.1 to 10,000 parasites/mL or DNA from clinical samples (56 AF samples, 55 placentas, and various other samples, of which 95 were from patients with proven toxoplasmosis) were tested. DNA extractions performed with the EXTRAblood or the QIAamp DNA-minikit were compared. The ELITe MGB assay amplified less frequently low-concentration replicates (less than 10 parasites/mL) of calibrated suspensions than the reference French laboratories. The combination of EXTRAblood and Toxoplasma ELITe MGB yielded poorer sensitivity than the combination of QIAmp DNA-minikit and ELITe MGB for low parasite concentrations (1 parasite/mL). The Toxoplasma ELITe MGB sensitivity ranged from 79% to 100% when DNA from placenta and AF samples was tested. Overall, the authors concluded that the Toxoplasma ELITe MGB assay associated with the QIAamp DNA-minikit is suitable for the diagnosis of toxoplasmosis from noncell-rich or nonhemoglobin-rich samples, which is the case of AF samples.
Regarding the treatment of acquired toxoplasmosis during pregnancy, in Europe and in the United States, the recommendation is for a spiramycin prescription when seroconversion of pregnant women is detected. The exchange to a new therapeutic regimen composed of sulfadiazine, pyrimethamine, and weekly supplements of folinic acid after 18 weeks of gestation is proposed when PCR in AF sample is positive or there are changes in fetal ultrasonography. To date, there is no unequivocal evidence of toxoplasmosis treatment efficacy during pregnancy as a result of the lack of randomized controlled studies.28
The Brazilian Ministry of Health recommends spiramycin prescription, which is freely distributed to patients, as soon as the maternal serologic diagnosis is established, before amniocentesis.12 All of the pregnant women in this study had positive conventional PCR and quantitative PCR in AF samples and received spiramycin. However, none of them had the therapeutic regimen changed, even in the presence of positive PCR and fetal ultrasound abnormalities. The justification for the maintenance of spiramycin was that patients could not afford pyrimethamine, sulfadiazine, and leucovorin, which are not provided by the government as is spiramycin.
A Brazilian report has found a reduced transmission of T gondii in spiramycin-treated pregnant women, justifying the maintenance of spiramycin in this study.29 More recently, Avci et al11 corroborated the efficacy of spiramycin to treat pregnant women with anti–T gondii IgM positivity who had low IgG avidity indices within the first trimester. These patients had amniocentesis performed at 19–21 weeks of gestation and AF samples examined for the detection of T gondii DNA. Detailed ultrasonographic examinations were performed between 20 and 24 weeks of gestation. Mothers and infants were monitored for at least 1 year. Results showed that 55 of 61 (90.2%) patients had received spiramycin prophylaxis, whereas six (9.8%) women refused treatment. Polymerase chain reaction was positive in AF samples of four (6.6%) women, all of them belonging to the group without treatment.
The pregnant women in this study received 3 g/d of spiramycin and this represents a confounding factor for the interpretation of laboratory results such as serology29 and quantitative PCR. Regarding the possible interference of spiramycin on parasite load levels, according to Schoondermark-Van De Ven et al,30 only in amniocentesis performed before 3 weeks of treatment (there were only four participants in this study), the “steady state” of the drug might not have been attained. On the contrary, Gratzl et al31 measured spiramycin concentrations in 18 AF samples obtained 5–109 days after prescription of spiramycin, and none of the concentrations were within the range to inhibit the parasite growth in vitro, so that, according to these authors, even in the pregnant women punctured after 4 weeks of treatment (most participants in this study), the parasites should be still replicating.
The present study did not include women who acquired toxoplasmosis during the third trimester of pregnancy as a result of the inability to have access to these patients who received medical care in another outpatient clinic. Third-trimester infections are more common than those of the first and second trimesters and should have been studied, although they are less deleterious to fetuses.1,2
In conclusion, the presence of signs and severity of congenital toxoplasmosis was associated with parasite load levels, irrespective of gestational age at the time of maternal infection. More importantly, the distribution of the parasite load levels by percentile has demonstrated that symptomatic patients appeared from the 75th percentile and severely compromised patients from the 95th percentile. The parasite load showed a double OR for each 20-parasite/mL increment, and the AUC of parasite load alone or combined with the one of gestational age at maternal infection was also associated with the infant's clinical outcome. A laboratory tool that can estimate the risk for symptomatic and severe congenital toxoplasmosis, independently of gestational age at maternal infection, constitutes a very promising prognosis marker that may be relevant for the clinical management of congenital toxoplasmosis. In clinical practice, fetuses with high parasite load in AF could receive more aggressive treatment regimens from the beginning in addition to clinical, ultrasonography, and laboratory examinations performed more frequently.
1. Dubey JP, Lago EG, Gennari SM, Su C, Jones JL. Toxoplasmosis in humans and animals in Brazil: high prevalence, high burden of disease, and epidemiology. Parasitology 2012;139:1375–424.
2. Robert-Gangneux F. It is not only the cat that did it: how to prevent and treat congenital toxoplasmosis. J Infect 2014;68(suppl 1):S125–33.
3. Costa JM, Ernault P, Gautier E, Bretagne S. Prenatal diagnosis of congenital toxoplasmosis by duplex real-time PCR using fluorescence resonance energy transfer hybridization probes. Prenat Diagn 2001;21:85–8.
4. Romand S, Chosson M, Franck J, Wallon M, Kieffer F, Kaiser K, et al. Usefulness of quantitative polymerase chain reaction in amniotic fluid as early prognostic marker of fetal infection with Toxoplasma gondii. Am J Obstet Gynecol 2004;190:797–802.
5. Tan TG, Mui E, Cong H, Witola WH, Montpetit A, Muench SP, et al. Identification of T. gondii epitopes, adjuvants, and host genetic factors that influence protection of mice and humans. Vaccine 2010;28:3977–89.
6. Wallon M, Franck J, Thulliez P, Huissoud C, Peyron F, Garcia-Meric P, et al. Accuracy of real-time polymerase chain reaction for Toxoplasma gondii in amniotic fluid. Obstet Gynecol 2010;115:727–33.
7. Shimokawa PT, Targa LS, Yamamoto L, Rodrigues JC, Kanunfre KA, Okay TS. HLA-DQA1/B1 alleles as putative susceptibility markers in congenital toxoplasmosis. Virulence 2016;7:456–64.
8. Teixeira LE, Kanunfre KA, Shimokawa PT, Targa LS, Rodrigues JC, Domingues W, et al. The performance of four molecular methods for the laboratory diagnosis of congenital toxoplasmosis in amniotic fluid samples. Rev Soc Bras Med Trop 2013;46:584–8.
9. Costa JG, Carneiro AC, Tavares AT, Andrade GM, Vasconcelos-Santos DV, Januário JN, et al. Real-time PCR as a prognostic tool for human congenital toxoplasmosis. J Clin Microbiol 2013;51:2766–8.
10. Camargo ME. Improved technique of indirect immunofluorescence for serological diagnosis of toxoplasmosis. Rev Inst Med Trop Sao Paulo 1964;6:117–8.
11. Avci ME, Arslan F, Çiftçi Ş, Ekiz A, Tüten A, Yildirim G, et al. Role of spiramycin in prevention of fetal toxoplasmosis. J Matern Fetal Neonatal Med 2016;29:2073–6.
12. Brazilian Ministry of Health. High-risk pregnancy: technical manual. 5th ed. Brasília (Ministério da Saúde): Institutional publication; 2012:115–118.
13. Pomares C, Montoya JG. Laboratory diagnosis of congenital toxoplasmosis. J Clin Microbiol 2016;54:2448–54.
14. Bayley N. Bayley scales of infant development. 2nd ed. San Antonio (TX): Psychological Corporation; 1993.
15. McAuley J, Boyer KM, Patel D, Mets M, Swisher C, Roizen N, et al. Early and longitudinal evaluations of treated infants and children and untreated historical patients with congenital toxoplasmosis: the Chicago Collaborative Treatment Trial. Clin Infect Dis 1994;18:38–72.
16. Petersen E. Toxoplasmosis. Semin Fetal Neonatal Med 2007;12:214–23.
17. Okay TS, Yamamoto L, Oliveira LC, Manuli ER, Andrade Junior HF, Del Negro GM. Significant performance variation among PCR systems in diagnosing congenital toxoplasmosis in São Paulo, Brazil: analysis of 467 amniotic fluid samples. Clinics (Sao Paulo) 2009;64:171–6.
18. Burg JL, Grover CM, Pouletty P, Boothroyd JC. Direct and sensitive detection of a pathogenic protozoan, Toxoplasma gondii, by polymerase chain reaction. J Clin Microbiol 1989;27:1787–92.
19. Pereira AP, Dias MA, Bastos MH, da Gama SG, Leal Mdo C. Determining gestational age for public health care users in Brazil: comparison of methods and algorithm creation. BMC Res Notes 2013;6:60.
20. Alexander GR, Himes JH, Kaufman RB, Mor J, Kogan M. A United States national reference for fetal growth. Obstet Gynecol 1996;87:163–8.
21. Sabin AB. Toxoplasmic encephalitis in children. JAMA 1941;116:801–7.
22. Robert-Gangneux F, Brenier-Pinchart MP, Yera H, Belaz S, Varlet-Marie E, Bastien P, et al. Evaluation of the Toxoplasma ELITe MGB Real-Time PCR assay for the diagnosis of toxoplasmosis. J Clin Microbiol 2017;55:1369–1376.
23. Wallon M, Peyron F, Cornu C, Vinault S, Abrahamowicz M, Kopp CB, et al. Congenital toxoplasma infection: monthly prenatal screening decreases transmission rate and improves clinical outcome at age 3 years. Clin Infect Dis 2013;56:1223–31.
24. Daffos F, Forestier F, Capella-Pavlovsky M, Thulliez P, Aufrant C, Valenti D, et al. Prenatal management of 746 pregnancies at risk for congenital toxoplasmosis. N Engl J Med 1988;318:271–5.
25. Desmonts G, Couvreur J. Congenital toxoplasmosis. A prospective study of 378 pregnancies. N Engl J Med 1974;290:1110–6.
26. Dunn D, Wallon M, Peyron F, Petersen E, Peckham C, Gilbert R. Mother-to-child transmission of toxoplasmosis: risk estimates for clinical counselling. Lancet 1999;353:1829–33.
27. Homan WL, Vercammen M, De Braekeleer J, Verschueren H. Identification of a 200- to 300-fold repetitive 529 bp DNA fragment in Toxoplasma gondii, and its use for diagnostic and quantitative PCR. Int J Parasitol 2000;30:69–75.
28. Peyron F. When are we going to celebrate the centenary of the discovery of efficient treatment for congenital toxoplasmosis? Mem Inst Oswaldo Cruz 2009;104:316–9.
29. Rodrigues IM, Costa TL, Avelar JB, Amaral WN, Castro AM, Avelino MM. Assessment of laboratory methods used in the diagnosis of congenital toxoplasmosis after maternal treatment with spiramycin in pregnancy. BMC Infect Dis 2014;14:349.
30. Schoondermark-Van de Ven E, Melchers W, Camps W, Eskes T, Meuwissen J, Galama J. Effectiveness of spiramycin for treatment of congenital Toxoplasma gondii infection in rhesus monkeys. Antimicrob Agents Chemother 1994;38:1930–6.
© 2017 by The American College of Obstetricians and Gynecologists. Published by Wolters Kluwer Health, Inc. All rights reserved.
31. Gratzl R, Sodeck G, Platzer P, Jäger W, Graf J, Pollak A, et al. Treatment of toxoplasmosis in pregnancy: concentrations of spiramycin and neospiramycin in maternal serum and amniotic fluid. Eur J Clin Microbiol Infect Dis 2002;21:12–6.