Abstract

Objective. To understand how congenital toxoplasmosis (CT) diagnosis has evolved over the years, we performed a systematic review and meta-analysis to summarize the kind of analysis that has been employed for CT diagnosis. Methods. PubMed and Lilacs databases were used in order to access the kind of analysis that has been employed for CT diagnosis in several samples. Our search combined the following combining terms: “congenital toxoplasmosis” or “gestational toxoplasmosis” and “diagnosis” and “blood,” “serum,” “amniotic fluid,” “placenta,” or “colostrum.” We extracted data on true positive, true negative, false positive, and false negative to generate pooled sensitivity, specificity, and diagnostic odds ratio (DOR). Random-effects models using MetaDTA were used for analysis. Results. Sixty-five articles were included in the study aiming for comparisons (75.4%), diagnosis performance (52.3%), diagnosis improvement (32.3%), or to distinguish acute/chronic infection phases (36.9%). Amniotic fluid (AF) and placenta were used in 36.9% and 10.8% of articles, respectively, targeting parasites and/or T. gondii DNA. Blood was used in 86% of articles for enzymatic assays. Colostrum was used in one article to search for antibodies. In meta-analysis, PCR in AF showed the best performance for CT diagnosis based on the highest summary sensitivity (85.1%) and specificity (99.7%) added to lower magnitude heterogeneity. Conclusion. Most of the assays being researched to diagnose CT are basically the same traditional approaches available for clinical purposes. The range in diagnostic performance and the challenges imposed by CT diagnosis indicate the need to better explore pregnancy samples in search of new possibilities for diagnostic tools. Exploring immunological markers and using bioinformatics tools and T. gondii recombinant antigens should address the research needed for a new generation of diagnostic tools to face these challenges.

1. Introduction

Congenital toxoplasmosis (CT) is a severe form of the disease caused by Toxoplasma gondii and occurs through the transplacental passage of tachyzoites from pregnant women to the fetus [1]. The risk of transmission depends on gestational age and clinical management for effective therapeutic intervention [2]. Infected fetuses and newborns can suffer serious consequences of infection, such as retinochoroiditis, encephalitis, intracranial calcification, hydrocephalus, and death [3].

Effective control and treatment of CT depend on accurate detection of T. gondii infection. The utilization of highly sensitive and specific diagnostic methods followed by treatment can prevent placental transmission, fetal infection, and sequelae to the fetus [4]. The laboratory diagnosis approaches commonly employed in CT are based on molecular, parasitological, and immunological assays such as PCR, bioassays, and immunoenzymatic assays, respectively. These methods allow detecting the parasite or antibodies using different samples such as amniotic fluid (AF) [523], umbilical cord blood, maternal and newborn blood [14–69], placental fragments [6268], and colostrum [69]. There are still numerous gaps in CT diagnosis. The variability in diagnostic performance, adding to the difficulty in interpreting results to differentiate the infection stage in pregnant women, causes delays in diagnosis and treatment. Diagnostic failures are also associated with unnecessary amniocentesis or poorly designed treatment. All these conditions compromise gestational safety.

Procedures adopted to diagnose the infection, including test type/platform and antenatal period, vary according to the guidelines of each country/society. For example, in Brazil [70], CT is confirmed when a suspected case presents one of the following situations: presence of T. gondii DNA in AF, fetal tissue, or child body fluids; IgM or IgA and anti-T. gondii IgG reagent up to six months of life; serum levels of anti-T. gondii on the rise in at least two serial samples with a minimum interval of 3 weeks during the first 12 months of life; anti-T. gondii IgG persistently reactive after 12 months of age; retinochoroiditis, hydrocephalus, or cerebral calcification (or associations between the signs) with reactive IgG.

Several countries have CT surveillance programs, but robust information on the frequency of CT transmission is limited to a few countries [71], so CT is substantially underestimated worldwide [72]. Despite this, published data show that T. gondii is responsible for almost two-thirds of the estimated 1.9 million disability-adjusted life years (DALYs) [73], with an estimated 190,000 cases annually [74]. The incidence estimation of CT can be obtained from case report series, inferences from gestational toxoplasmosis, and testing babies at birth [71]. The disease is associated with fetal loss and neonatal death in approximately 3% of cases, [75] as well as craniocerebral/ocular sequelae [76]. Subclinical disease at birth is present in 75% of cases, with symptoms that may start many years or even decades later [74].

To provide an understanding of the evolution of CT diagnosis over the years, we present here a review of methods that are currently employed for prenatal and postnatal CT diagnosis in several samples. It emphasizes the sample type, targets, and methods applied to diagnosis at different gestational ages using biological samples from pregnant women, fetuses, and newborns. Moreover, it brings insights into possible future challenges of CT diagnosis.

2. Methods

2.1. Search Strategy, Study Selection, and Data Extraction

Our study followed the preferred reporting items for systematic reviews (PRISMA) guidelines [46]. PubMed and Lilacs citation databases were searched from 2001 to 2020, combining the terms “congenital toxoplasmosis” or “gestational toxoplasmosis” and “diagnosis” and “blood” or “serum” or “amniotic fluid” or “placenta” or “colostrum.” Only papers using human samples and written in English were included.

The articles were selected by the Rayyan program, and seven authors conducted the preliminary selection based on abstracts and paper titles. After the first selection, conflicting decisions by at least three authors were considered for a second blind analysis. Afterward, articles considered eligible by at least four authors were included in the preliminary screening for full reading. Studies with at least one of the following criteria were excluded: studies evaluating exclusively infant samples, reviews or descriptive studies, articles with no eligible data, case reports, and studies approaching multiple infections.

The following data were recorded from the selected studies: major goals, sample type, gestational age at sample collection, laboratory methods, and major results. For meta-analysis, data on molecular (PCR) and bioassay diagnosis performance were collected from articles that included these analyses and provided data about the number of samples, sensitivity, and specificity. Only studies that reported the true positive, false negative, true negative, and false positive values or that these values could be calculated were included. Investigators collected data independently. When literature data interpretation was controversial, investigators discussed it and reached a consensus. Some studies considered essential to the review that were not included in any of the research bases were added to the introduction and discussion.

2.2. Statistical Analysis

Venn diagram was performed using Bioinformatics and Evolutionary Genomics, available at https://bioinformatics.psb.ugent.be/webtools/Venn/. It was constructed to identify common and exclusive biological samples used in the selected studies. Meta-analysis of molecular diagnosis or bioassay in AF or placenta samples was performed using MetaDTA (version 2.0) [77, 78], available at https://crsu.shinyapps.io/dta_ma/. The diagnostic odds ratio (DOR), positive likelihood (LR+), and negative likelihood ratio (LR−) were used to determine the overall diagnostic accuracy. Sensitivity and specificity points were shown along with forest plots and SROC curves. The forest plots were edited using GraphPad Prism software. Heterogeneity and threshold effects were evaluated using the visual summary of SROC plots and random effects correlation, as described by Druce et al. [79]. All summary parameters were calculated along with the associated 95% confidence interval (CI).

3. Results

3.1. Analysis of the Included Literature

Through this systematic review, 1137 articles were found following the initial database search. In total, 517 articles were excluded from duplicate records, and 620 articles were screened based on title and abstract. From that, 523 articles were excluded, as it did not fit our filters. The remaining 97 articles were evaluated by full reading, and 32 were excluded according to the criteria outlined in the Methods. Finally, 65 articles were included in the systematic review, and of these, 10 articles aiming for diagnostic performance on molecular assays (PCR) and/or bioassay were selected for meta-analysis. Details of the search and study selection procedures were described in a PRISMA flow diagram (Figure 1(a)).

To identify common and exclusive samples used in the selected studies, a Venn diagram was constructed (Figure 1(b)). The analysis of samples used in the articles demonstrated that 9 studies were conducted employing exclusively AF, whereas 38 studies were performed exclusively with blood samples. Analyzing the articles with more than one sample, it was demonstrated that 10 articles used AF and blood, 2 articles used placenta and blood, 1 article used colostrum and blood, and 5 articles used AF, blood, and placenta simultaneously (Figure 1(b)).

3.2. Major Goals of the Selected Articles

The major goals explored in the 65 selected articles were analyzed (Table 1), and 34 of them aimed to analyze diagnosis performance. Fifteen articles presented diagnosis performance on molecular diagnosis and/or bioassay and 23 articles analyzed serological performance diagnosis. Twenty-two articles in this review aimed at diagnosis improvement, 9 of them using a combination of diagnosis assays to improve performance [7, 14, 15, 21, 30, 6264, 69] and 12 employing some modification of available methodologies [6, 27, 32, 34, 39, 4244, 46, 47, 56, 60].

Most of the articles in this review, 49 articles, aimed comparisons: performance comparisons based on time of sample collection [5, 13, 18, 22, 46], comparisons between assays [6–69], comparisons between samples [18, 20, 21, 65, 68, 69], and other comparisons [5, 22, 53].

Twenty-four articles aimed to distinguish acute/chronic phases of infection, and all of them employed blood samples [21, 2427, 29, 31, 33, 36, 3841, 44, 47, 51, 52, 5457, 5961]. Twenty-four articles presented other objectives, such as evaluating the treatment effect on CT diagnosis [5, 13, 27, 64], correlating parasite load to CT severity [6, 22], characterizing T. gondii strains on CT [12, 16], and others [17, 19, 23, 28, 38, 39, 42, 4850, 58, 59, 62, 63, 67, 69].

3.3. Diagnostic Methods That Employed Amniotic Fluid Samples

From the careful selection, 24 articles from the total used AF as a sample (Table 2). Regarding gestational age on sample collection, 5 articles collected AF between the 14th and 26th gestational weeks (GW), 9 articles collected AF between the 14th and 41st GW, and 5 articles collected additional samples at birth. Ten articles did not provide details on the date of sample collection.

All selected articles used parasites and/or T. gondii DNA as targets of study. Concerning the assays performed in those studies, all selected articles performed PCR and 8 of them also performed bioassay by mouse inoculation. The B1 gene was the most commonly used gene in PCR (19/24 articles). Twelve articles (12/24 articles) used other genes such as 529-bp, RE-sequence, P30, and others. Two articles (2/24 articles) did not provide details about PCR. Most studies using bioassay did not provide information about the methodology employed.

3.4. Diagnostic Methods That Employed Blood Sample

Of the total, 56 articles used blood samples for CT diagnosis (Table 3). Peripheral blood samples were collected from pregnant women (M-PB) (51 articles/78.4%), cord blood by cordocentesis (P-CB) (2 articles), cord blood at the time of delivery (N-CB) (11 articles), or/and neonatal peripheral blood (N-PB) (18 articles).

When the target was examined, 53 of 56 articles analyzed antibodies against T. gondii by immunoassays. The performed serological methods were enzyme assays (ELISA, VIDAS, Enzygnost, Platelia, AxSYM, Cobas, EIA, WB, MEIA, and ELIFA) in 49 of 53 articles; agglutination (ISAGA, DA, HSDA, ICT, AC/HS) in 25 articles; fluorescence (IFAT, IMX, IF, FAT, ELFA, and FEIA) in 23 articles; and chemiluminescence (Architect, ECLIA, Liaison, CML, Vidia, and Advia Centaur) in 6 articles. The Sabin–Feldman dye test (SFDT) was used in 9 articles, and latex agglutination test, laser immunonephelometry, or lateral flow immunoassay (LFIA) were used in 1 article.

Analysis of parasite and/or T. gondii DNA in blood samples was applied in 12 of 56 articles. For parasite/DNA detection, 11 articles used PCR and 2 articles used bioassay. The B1 gene was the most commonly used in PCR.

Regarding the type of blood samples and assay employed for diagnosis, all articles with M-PB (n = 51) used serological methods. IgM and IgG were the most assessed molecules (IgM: 47 articles, IgG: 51, IgA: 9, and IgE: 2). IgG avidity was analyzed in 32 articles and IgG subclasses in 1 article. Enzyme assays were performed in 47 articles, agglutination assays in 19 articles, fluorescence assays in 21 articles, and chemiluminescence assays in 8 articles. Eight articles used M-PB to assess T. gondii DNA by PCR.

All articles with P-CB (n = 2) used serological methods to analyze antibodies (IgM: 2 articles, IgA: 2, IgG: 1). No article analyzed IgG subclasses or IgE. Agglutination assays were performed in all articles, and enzyme or fluorescence assays were performed in 1 article. One article used P-CB samples for bioassay, but PCR was not performed.

Nine of 11 articles used serological methods to analyze N-CB. IgM and IgG antibodies were the most assessed molecules (IgM: 9 articles, IgG: 5, IgA: 2). No article analyzed IgG avidity, IgG subclasses, or IgE. The enzyme and agglutination assays were performed in 4 articles and the fluorescence assays in 2. Four of 11 articles used N-CB to assess T. gondii DNA by PCR and 2 for bioassay.

Seventeen of 18 articles used serological methods to analyze N-PB. IgM and IgG were the most assessed molecules (IgM: 16 articles, IgG: 15, IgA: 5). No article has analyzed IgE. IgG avidity was analyzed in 5 articles and IgG subclasses in 1 article. Enzyme assays were performed in 14 articles, agglutination assays in 13, fluorescence assays in 8, and chemiluminescence assays in 2 articles. One article used N-PB to assess T. gondii DNA by PCR. No article employed N-PB to perform bioassay.

3.5. Diagnostic Methods That Employed Placenta and Colostrum Samples

Analyzing articles that employed samples for postnatal diagnosis, it was detected that 7 articles used placenta and 1 used colostrum for CT diagnosis (Table 4). Placentas were used to search for parasites and/or T. gondii DNA by PCR or bioassay. PCR and bioassay were performed in 4 studies, whereas 2 articles used exclusively PCR and 1 article used exclusively bioassay. B1 was the most commonly used gene for PCR in placentas. REP529 and RE-sequence were applied in the PCR in 2 and 1 articles, respectively. Concerning the bioassay, 2 articles employed Swiss females and 2 other articles did not report details about this methodology. Colostrum was collected up to 3 days after birth. Samples were analyzed to detect anti-T. gondii antibodies using ELISA and western blot immunoassays.

3.6. Measures of Diagnostic Performance

From the 10 articles included in the meta-analysis, 8 used AF for PCR [5, 7, 11, 14, 19, 20, 66, 67] and 4 used AF for bioassay [7, 14, 66, 67]. Four articles used placenta for both PCR and bioassay techniques [62, 63, 66, 67]. The estimated sensitivity and specificity of PCR in AF were 85.1% (95% CI 69.5–94.4%) and 99.7% (95% CI 97.2−1.00%), respectively. The sensitivity and specificity of the bioassay in AF were 75.4% (95% CI 41.6–71.8%) and 99.3% (95% CI 93.6–99.9%), respectively. PCR in placenta had an estimated sensitivity of 58.9% (95% CI 58.5–59.3%) and a specificity of 96.3% (95% CI 96.3–96.4%). Bioassay in placenta had an estimated sensitivity of 58.6% (95% CI 47.2–69.2%) and a specificity of 99.5% (95% CI 97.9–99.9%). Paired forest plots are shown in Figure 2 and Table 5. The RE correlation for the bioassay presented values of +1 for AF and −1 for PL. For the PCR scenarios, the RE correlation values were −0.572 for PCR in AF and −0.365 in PL (Table 5).

The odds ratios determined by PCR or bioassay of AF and placenta were combined for quantitative comparison. DOR was 2018.385 (95% CI 228.652–17816.960) for PCR in AF, 189.94 (95% CI 13.45–2681.75) for bioassay in AF, 37.70 (95% CI 36.8–38.58) for PCR in placenta, and 258.86 (95% CI 69.77–960.40) for bioassay in placenta (Table 4).

PCR in AF showed higher LR+ (302.048 with 95% CI 30.916–2950.945) and lower LR− (0.150 with 95% CI: 0.069–0.325) compared to the other techniques (Table 5).

4. Discussion

To the best of our knowledge, the first conclusive reported case of toxoplasmosis in newborns was diagnosed based on encephalomyelitis and chorioretinitis findings in infant postmortem tissues. Mice and rabbit tissue inoculations evidenced an infection with protozoa morphologically compatible with T. gondii. [80] Afterward, a dye test was developed to evaluate the presence of the specific antibody [81]. Following, a description of T. gondii isolation from the placenta gave a new puzzle connection about congenital infection [82]. Later, a 15-year prospective study brought important information about CT diagnosis [83].

The present study aimed to investigate commonly used diagnostic methods for CT and understand the accuracy of these methodologies. From these data, we seek out new diagnostic proposals that can be investigated, bringing insights into new diagnostic approaches. Our data suggested that, in the last 20 years, the samples and assays used for CT diagnosis are basically the same as those of past decades. Few studies evaluated the effectiveness of alternative samples, such as colostrum. The majority of the studies in this review used blood samples mainly for serological screening, and a few studies used more than one type of sample for diagnostic investigation. The increase in amount and time of sample collection represents a gain for CT diagnosis that has also evolved in accuracy [30]. A schematic model representing the types of samples and methods used for T. gondii detection before and after birth is shown in Figure 3.

Techniques used for CT diagnosis have advanced; however, many difficulties are still encountered in screening pregnant women and fetuses. Our review suggested that one of the major challenges of CT diagnosis is dating the T. gondii infection [30]. T. gondii-specific immunoglobulin (IgG and IgM) searches are often used to investigate when the infection occurred [3, 25, 30]. IgG-avidity helps determine the risk of T. gondii transmission at any time during pregnancy [29]. Conversely, avidity assay results classified as borderline or low can be erroneously interpreted as consistent with a recently acquired infection [25].

Many studies in this review aimed to distinguish between acute/chronic infection phases. However, few of them used new approaches, such as evaluating more options for T. gondii antigens for the improvement of enzyme assays [26, 33, 39, 44, 47] or employing bioinformatics tools, such as epitope prediction for CT diagnosis [56]. The use of specific molecular markers is a promising option in T. gondii serodiagnosis and can be useful for dating the infection. Recombinant proteins are highly advantageous for improving the diagnostic assay. The combination of several recombinant antigens with multiple immunodominant epitopes significantly increases the probability of detecting specific antibodies at different stages of the infection [4, 84]. Besides, avidity assays based on recombinant antigens have potential clinical usefulness for diagnosing the acute phase of T. gondii infection [26].

Many articles in this review assessed performance diagnosis or combined methods to evaluate performance improvement. The importance of diagnostic accuracy should be emphasized in order to conduct the correct treatment to avoid transplacental transmission and to prevent unnecessary and potentially toxic treatment or termination of pregnancy [5]. A combination of methods can also improve diagnostic accuracy [85]. T. gondii detection by DNA amplification or parasite isolation is complementary to serological tests. These methods are particularly important in AF to indicate fetal infection [86].

Our meta-analysis of diagnostic performance for DNA/parasite detection (PCR and bioassay) demonstrated a variation of sensitivity values. Although amniocentesis is a highly invasive method, amniotic fluid was the sample that presented the best values of PCR sensitivity. Likewise, detecting parasite burden in AF helps predict the severity of clinical symptoms in neonates congenitally infected [72]. The antenatal diagnosis of CT is the greatest advance in the cases of fetal infection, and the use of PCR analysis of AF is the most commonly used and accepted laboratory method for CT diagnosis during gestation [86]. Normally, negative PCR in AF indicates the absence of fetal infection, although it cannot be ruled out completely. However, a positive PCR result almost certainly indicates a congenital infection [14].

Variation in sensitivity can be associated with dissimilarity in the PCR methodologies, time of sample collection, time of maternal seroconversion, influence of treatment [19], and disparity in performance among laboratories. A considerable number of PCR results show the absence of T. gondii DNA amplification concomitant with CT, indicating low sensitivity [13]. In some cases, these results can be explained by the absence of optimal amniocentesis at the time of sample collection [23]. It can also be attributed to the inefficiency of parasite DNA extraction and amplification, mainly due to the low concentrations of tachyzoites in the AF collected [87].

Some studies in this review implemented comparisons of the sensitivities of PCR methodologies, including comparisons between target genes. By far, B1 is the most employed gene in PCR followed by REP529. There is an important discrepancy in the literature concerning the best target gene for PCRs. Some studies in this review indicated that REP529 is more sensitive compared to B1 [18, 68]. Another study showed no discordance between these two targets [88]. These results drive the need for more studies comparing target genes.

The present study aimed to estimate all possible random effects for CT diagnosis data and compare them without applying alternative simplifications [89]. Random effect correlations of +1 or −1 found for bioassay indicated a hit/truncation on model parameters. There were no explicit convergence problems, yet a few studies and/or sparse data (e.g., indicating no/low heterogeneity in the specificity parameter) are possibly data problems. In this situation, the power of the model can be compromised [90] and the diagnostic parameters should be interpreted with caution.

Our study to review pertinent publications and assess diagnostic test accuracy performance for detecting T. gondii infection was defined by the limited number of available studies that meet the selection criteria. Insufficient reporting regarding population characteristics/recruitment and data about sample number, sensitivity, and specificity was an issue in many studies, with information often provided with little detail.

Our results also indicate the need for searching for new diagnosis methodologies, improving existing techniques, and providing proper training for professionals involved in the routine diagnosis of CT [91]. Moreover, since higher T. gondii concentrations in AF are correlated with clinical signs in neonates, quantitative PCR can be important to evaluate the prognosis of the fetal infection [92]. Another important methodology that was scarcely used in selected articles, but is of great relevance, is T. gondii isolation followed by genotyping. This technique allows the identification of nonclonal strains. Dubey and coworkers [93] identified 58 different T. gondii genotypes circulating in Brazil. Studies identified a new T. gondii strain in southern Brazil that was strongly related to the toxoplasmosis outbreak in 2018 [94]. Such studies allow the surveillance of new circulating genotypes that are usually related to more severe forms of toxoplasmosis.

Variation in performance in diagnostics also drives the need for new diagnostic approaches, and few studies have focused on this aim. There is a great potential for using biomolecules present in AF for complementary diagnosis. Using complementary biomarkers, such as immune response mediators, could help endorse and increase the reliability of diagnosis in AF. Previous studies suggested the ability to use cytokines, such as TGF-β in AF, as biomarkers to predict acute T. gondii infection [95]. Another possibility is the use of AF cellularity as a potential biomarker of congenital infection. Studies using AF from the second and third trimester of pregnancy highlighted the cell dynamics in this compartment [96]. Inflammation, whether associated or not with infection, causes an increase in the number of immune cells.

Some studies in this review collected samples at birth, including AF, placenta, and blood. Samples collected at the time of delivery are especially significant in the absence of prenatal follow-up, making it possible to anticipate diagnosis and treatment for newborns. T. gondii isolation from placenta is a useful tool to study CT and is an easily available sample. Placental analysis can be important to diagnose infection when AF is either not positive or not analyzed. Besides, placental samples can be useful for isolation and genotyping of the parasite, especially in outbreaks [97]. However, it is important to highlight that maternal treatment can influence the efficacy of placental analysis since T. gondii was less frequently isolated in the placenta of treated women [62, 64, 66].

Postnatal follow-up, based on after-birth samples, remains necessary in the first year of life to fully exclude the infection when PCR or serological results were negative [29, 86]. Diagnosis based on cell immunity has been increasingly used as a complementary diagnostic to monitoring infants [98, 99]. This potential methodology should also be explored for maternal samples during prenatal follow-up. Alternative biological samples, such as colostrum from puerperal women [69] and saliva [44, 100], also provide interesting data on humoral immunity and promising results for diagnosing toxoplasmosis using noninvasive sample collection. Searching for IgG and IgG-subclasses produced by newborns compared with maternal antibody responses [32] can also be promising for CT diagnosis.

Preventive and diagnostic measures for pregnant women vary between countries. Although prenatal diagnosis of CT is available, there is no international framework for monitoring the disease, and it is a neglected disease in most countries [1]. The absence or incomplete prenatal screening and treatment have been identified as an important risk factor for CT [87]. Thus, the screening and prevention measures against toxoplasmosis should be made mandatory for pregnant women attending the antenatal clinic [4].

5. Conclusion

In conclusion, this review points out that the assays employed in the research are basically the same traditional approaches available for clinical purposes. These assays showed important variations in diagnostic performance that can result in undiagnosed CT. These results challenge us to search for new generations of diagnostic tools and improve existing techniques, together with efforts towards increasing the feasibility of laboratory testing.

Data Availability

The data used to support the findings of this systematic review are available in the References section of the article.

Additional Points

What is Already Known about This Topic? (i) Advances in immunology, molecular biology, and bioinformatics have provided a new generation of diagnostic tools. (ii) CT diagnosis is essential for the treatment and clinical management of disease. (iii) New possibilities for diagnostic tools have been searched to improve the diagnosis of acquired toxoplasmosis. (iv) Pregnancy provides many potential samples to be used in CT diagnosis, presenting a wide array of new possibilities. What Does This Study Add? (i) Pregnancy samples have been poorly explored for searching the new generation of diagnostic tools. (ii) The assays used in the research to diagnose CT are basically the same traditional approaches available for clinical purposes. (iii) PCR and bioassay showed large variations in diagnostic performance. (iv) Improvement and searching for new forms of diagnosis are required for CT. (v) The search for immunological markers, the use of T. gondii recombinant antigens, and the identification of the genetic diversity of T. gondii strains are diagnostic possibilities to be explored using pregnancy samples.

Ethical Approval

This systematic review was performed using available databases and did not collect any personal or confidential information from participants.

Conflicts of Interest

The authors declare that they have no conflicts of interest.

Acknowledgments

The authors wish to thank the financial support received from the following Brazilian Research Agencies: Fundação de Amparo à Pesquisa de Minas Gerais (FAPEMIG), Conselho Nacional de Pesquisa Científica e Tecnológica (CNPq), and Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES).