Autism Spectrum Disorder (ASD) was described by Leo Kanner in 1943. According to DSM-5, ASD is a neurodevelopmental disorder in which there are persistent deficits in communication and social interaction, associated with restricted and repetitive patterns in behavior, interests, and activities [1]. Moreover, ASD encompasses a broad, heterogeneous manifestation that can be influenced by environmental and genetic factors. It is believed that the environment could be associated with prenatal and postnatal interactions, and in this review, we’ll focus on this subject [2].

It is known that infection and inflammation during pregnancy can have an important impact on fetal brain development, although the mechanisms of this harm are not fully clarified. Although TORCH infections (Toxoplasma, Rubella, Cytomegalovirus, Herpes simplex, and Syphilis) pathogens are directly related to teratogenicity, emerging studies suggest their associations with a wider spectrum of lesions to fetal neurodevelopment. Over the past thirty years, more evidence has been raised, which indicates that a wide range of infections during pregnancy could lead to a higher risk of neurodevelopmental disorders, such as ASD [3].

Thus, the aim of this systematic review (SR) is to study if there is a relationship between congenital infections (Toxoplasmosis, Cytomegalovirus, Rubella, Herpes simplex type 1 and 2, HIV, Zika and Syphilis) and ASD. Above all, the authors seek to identify loopholes in the current knowledge about this content and to understand the role of congenital infections in children’s development. Therefore, the authors hypothesize that environmental interaction between an infectious pathogen with a fetal organism in development, associated with a genetic susceptibility, can lead to neurodevelopmental disturbance, especially ASD.

This study is a systematic review. Hence, the following Mesh terms were used on databases from PubMED, Lilacs, and Embase, such as the following synonyms and acronyms: “Congenital Infection, Toxoplasma gondii” OR “CMV Infection, Congenital” OR “Rubella Syndrome OR “Congenital” OR “Congenital Zika Syndrome” OR “Virus, AIDS” OR “Syphilis, Congenital” AND “Autism Spectrum Disorders”. The search strategy was developed by an institutional librarian, reviewed by the senior researcher (MLN) and is available as supplementary material (Supplementary Table 1).

The inclusion criteria were children up to 7 years old with a diagnosis of congenital infection and confirmed ASD; articles written in English and published between 2014 - 2024, with the following designs (retrospective and prospective cohorts, longitudinal, cross-sectional, observational studies, case-control or case report). The exclusion criteria were articles in which the data presented did not allow ensuring how the diagnosis of congenital infection or ASD was based. This study was registered at PROSPERO (https://www.crd.york.ac.uk/prospero/) (ID CRD42024600301).

A total of four reviewers conducted the reading and selection of articles. The process was carried out in pairs, with each pair responsible for evaluating the inclusion and exclusion criteria. All articles retrieved from database searches were imported into Rayyan (https://rayyan.qcri.org), where duplicate articles were identified by the software and excluded after confirming titles and authors. Subsequently, the articles were screened by the four reviewers based on their title and abstract, with articles that did not address the relationship between one of the studied congenital infections and the development of ASD being excluded. After completing this stage, the reviewers performed a full-text reading of all selected studies for data extraction, which was carried out independently by the two pairs of reviewers.

The methodological quality of the included studies was assessed using the Newcastle-Ottawa Scale. Studies were classified according to their total score as follows: high quality, 7 to 9 points; moderate quality, 4 to 6 points; and low quality, 0 to 3 points. For analytical purposes, studies scoring 7 points or higher were considered to have good methodological quality, whereas studies scoring below 7 points were considered to have moderate or low methodological quality.

The extracted data were compiled into an Excel table with the following criteria: article characteristics (author identification, year of publication, journal name, DOI, title, objective, and study design) and sample characteristics (sample size, country of origin, follow-up period, and results). Discrepancies regarding the inclusion or exclusion of studies during this process were resolved through consensus discussions with a senior reviewer. To ensure the quality control of the selected articles, the Newcastle-Ottawa Scale (NOS) was applied, with scores ranging from 0 to 9, for case-control and cohort studies.

The present search yielded initially 286 articles, and after applying the inclusion criteria and performing a full-text review, 30 articles were included. To document the literature search process, the authors followed the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines. Figure 1 presents the flow diagram outlining the study selection process.

Fig. 1.

PRISMA flow chart showing the process of systematic article search and selection.

Table 1 summarizes the data extracted from the 30 articles included in this review. Results are presented according to the type of congenital infection. Of the included studies, 14 were case-control studies (mean NOS score = 6.7), 5 were prospective cohort studies (mean NOS score = 7.0), 7 were retrospective cohort studies (mean NOS score = 6.0), 3 were cross sectional and 1 case report.

A total of 10 articles addressed whether there is a relationship between ASD and Cytomegalovirus (CMV) [4–13]. One further study included CMV and toxoplasmosis patients [14]. These studies were published between 2015 and 2023. The design and quality of studies varied, and data was heterogeneous. Except for two studies that did not find a significant relationship between CMV and ASD, [13,14] the others suggested a higher prevalence of ASD in association of CMV congenital infection, some of them comparing the rates obtained to what was report regarding previous populational data [4,6,9].

Six articles studied ASD and the congenital Zika virus infection [15–20]. Two prospective and 2 retrospective cohorts, one case-control and one case report. The quality of studies varied, and data was heterogeneous.

Patients with Zika virus infection were at higher risk for developing neurodevelopmental abnormalities including ASD and its prevalence was higher than in the general population in some studies [15,16,20]. Also, more abnormalities in language, communication, fine and gross motor skills, cognitive and social-personal areas were reported [17,18]. In addition, there was a higher prevalence of male gender in the affected group [16]. Earlier exposition in pregnancy was associated with worse ASD features [17,18]. A case report supported that ASD can be a later manifestation of Zika virus infection even in children who do not have features of congenital Zika syndrome [19].

Six articles studied the relationship between ASD and Toxoplasmosis, which were published between 2016 and 2024, and one included mixed CMV and toxoplasmosis [14]. Six articles were case-control studies, [14,21–25] whose NOS scale ranged from 4 to 9, with a mean score of 7. Also, there was one cross-sectional study [26]. One case-control study found a significantly higher prevalence of IgG antibodies against T. gondii in the ASD group [25] and one cross-sectional study reported a higher prevalence of T. gondii infection in ASD children compared to a comparison group (p < 0.001) [26]. However, no clear association was observed in the others [14,21–24].

Three articles studied the relationship between ASD and Rubella [27–29] and were published between 2017 and 2023. A prospective cohort obtained an NOS score of 4 and a case-control of 6, while the last one was a cross-sectional study.

Patients with Congenital Rubella Syndrome had a higher risk for neurodevelopmental disability, particularly in communication and language domains [28] and children from mothers who seroconverted to rubella had a higher prevalence of an ASD outcome, however, the relationship did not achieve statistical significance [27].

In contrast, a small case-control study showed that individuals with seronegative of anti-rubella IgG may consider at risk for ASD while IgG seropositivity had a protective effect [29].

Two articles studied the relationship between ASD and Herpes. The studies were published in 2014 and 2018. One article was a case-control study [30] (NOS = 5) and another was a prospective cohort study (NOS = 7) [31]. None of them found a significant association between congenital herpes virus infection and ASD.

There was one article [32] that assessed whether there is a relationship between ASD and Varicella. The study was published in 2017, with a case-control design (NOS = 6). However, no significant association was identified.

There was one prospective cohort study (NOS = 7) that assessed the neurodevelopmental outcome in children who had Central Nervous System (CNS) infection by several pathogens and after evaluating the outcome for each causing agent Enterovirus appeared as another infectious agent associated to the development of ASD [33].

No articles regarding HIV and Syphilis were identified after applying the criteria for this review.

In this systematic review, the authors aimed to identify whether there is a relationship between congenital infections (Cytomegalovirus, Rubella, Herpes, Varicella, Zika, Toxoplasmosis, HIV, and Syphilis) and ASD. The data obtained were heterogeneous; however, 19 studies did find an association between congenital infection and the development of ASD and/or features of this spectrum. Furthermore, the present findings indicate that the link between congenital infections and ASD varies depending on the pathogen.

The majority of the studies suggest biological mechanisms of maternal immune activation, neuroinflammation, and genetic susceptibility as hypotheses that could explain a possible relationship between congenital infections and ASD. Infections during pregnancy trigger a strong maternal immune response, leading to an elevated level of pro-inflammatory cytokines that could affect the neurodevelopment of the fetus [26,33]. Epigenetic modifications were also addressed in some studies as possible mechanisms, suggesting that prenatal infections can generate long-term changes in gene expression related to synaptic function and neural connectivity. However, none of the studies were able to perfectly clarify the causality between infections and ASD, as well as the underlying mechanisms [4–12,15–20,24–28,30] Table 2 provides a summary of possible mechanisms that might explain the association of each congenital infection and the development of autism.

Regarding CMV infection, the majority of the studies have shown a possible association between this congenital infection and an increased risk of ASD, since they demonstrated a higher prevalence of CMV infection among children with ASD compared to a comparison group [4–10]. This congenital infection is known to induce a pro-inflammatory state in the fetal brain, characterized by increased levels of cytokines and chemokines, which can disrupt normal neurodevelopmental processes. The included studies suggest mechanisms such as direct injury to neural progenitor cells, disruption of neuronal migration and differentiation, and impairment of neural circuit formation during critical periods of neurodevelopment [4,9]. Furthermore, infection-induced neuroinflammation, characterized by microglial activation and increased production of pro-inflammatory cytokines, may interfere with synaptic pruning and the maturation of neural networks involved in social communication and behavior [11]. Evidence of white matter abnormalities, impaired brain connectivity, and hippocampal alterations in children with congenital CMV infection further supports the hypothesis that these structural and functional changes may contribute to ASD-related phenotypes [12,13]. Overall, current evidence supports an association with increased risk, likely mediated by interactions among infectious, immunological, and genetic susceptibility factors [5–8,10]. In a systematic review and meta-analyses, Maeyama and collaborators found a high prevalence of congenital CMV infection in ASD cases (OR 11.31, 95% CI 3.07–41.66). However, all three included studies had serious limitations [34].

The literature reviewed that addressed the relationship between congenital infection by Zika virus was more emphatic in the association, with ASD as one of the developmental disorders identified. In one study, it was suggested that the inflammatory response mediated by pro-inflammatory cytokines, triggered by ZIKV during pregnancy, may contribute to the development of neurodevelopmental disorders, such as ASD [19].

The studies that analyzed congenital Toxoplasmosis raised conflicting results, however, some studies reported a higher prevalence of toxoplasmosis infection in children with ASD [25,26]. Serological and molecular analyses suggested a positive relationship, where maternal toxoplasmosis infection was linked to mitochondrial DNA (mtDNA) and nuclear DNA (nDNA) damage, which may play a relevant role in autism development [24]. On the other hand, 2 studies suggested an inverse association, pointing out that higher maternal and neonatal IgG antibody levels were associated with a reduced risk of ASD, [23] and positive maternal IgM antibodies linked to lower odds of autism [21]. IgM secretion is one of the immune system’s initial responses to protect against a pathogen, and one interpretation is that a sufficient immune response to acute maternal T. gondii infection during pregnancy may be protective against ASD risk in offspring [21]. Thus, there may be a more complex pathophysiological mechanism involved, which requires further investigations.

Findings concerning the association between congenital Rubella infection and ASD were heterogeneous [27–29] and maybe an explanation may be the varied methodologies used. However, there are enough pathophysiological mechanisms that might link ASD and congenital rubella infection. One of the strongest theories suggests that the Rubivirus impairs the enzyme responsible for the conversion of retinol to retinoic acid and the catabolism of retinoic acid, resulting in retinoid toxicity. This process could disrupt normal fetal development, increasing the risk of ASD. However, the study emphasizes that, with the currently available evidence, it is still not possible to determine whether ASD is part of Congenital Rubella Syndrome. Therefore, further research is essential to clarify the role of Rubella in ASD.

The scarce data obtained in this review that evaluated a possible association between Herpes virus infection and ASD did not reach statistical significance. This finding was supported by previous literature [35] and the direct relationship between Herpes virus and ASD remains uncertain.

No pieces of evidence for an association between Varicella Zoster and ASD were obtained in this review. As only one single article, [32] with a small sample size, was retrieved in this review, more studies on this subject are necessary.

Moreover, other pathogens than the TORCH group can lead to neurodevelopmental impairment and ASD, such as Enterovirus [33] and it ought to trigger research into other infectious agents also.

Data obtained in this review allowed us to conclude that there is plausible evidence of a possible association between CMV, Zika, Toxoplasmosis, and Rubella infections and the development of ASD. Considering that congenital infections can disrupt neurogenesis and ASD has a multifactorial etiology, this is not an unexpected finding. Although there is no certainty of a common causal factor among the diseases, as their pathogenic mechanisms are not yet fully understood, this review highlights the possible positive association between congenital infections and ASD development, but there is an urgent need for more detailed studies and research on this topic.

Not applicable.

Not required.

Not applicable.

This work did not receive any specific financial support. MLN is a 1D researcher supported by the National Council for Scientific and Technological Development (CNPq)- Brazil, PQ 303,168/2021–8. RJKR and AWJ shared the PUCRS (BPA program) and CNPQ scholarship to develop this study.