Swedish researchers have suggested that a collection of rare infections passed from mothers to foetuses could increase the risk of autism or intellectual disability in their children, according to their population-based cohort study.
Among nearly 3.7m children born in Sweden, only 975 had congenital toxoplasmosis, syphilis, rubella, cytomegalovirus (CMV), or herpes simplex (TORCH) infections. Compared with uninfected children, however, those with TORCH infections were three times more likely to develop autism (adjusted HR 3.10) and seven times more likely to develop intellectual disabilities (adjusted HR 7.22).
MedPage Today reports those numbers suggest TORCH infections would be responsible for just 0.034% of the overall population's diagnosed autism cases – but roughly one in five infants exposed to a TORCH infection could be expected to develop autism, wrote Renee Gardner, PhD, of Karolinska Institutet in Stockholm, and colleagues in JAMA Paediatrics.
“Our findings suggest that TORCH infections may represent a rare example of perinatal environmental risk factors, providing evidence toward a consistent association with autism and intellectual disability,” they observed. “Associations were highly consistent between sibling and population-based models, indicating minimal confounding by shared familial factors.”
TORCH infections seem to be quite different from the more common infections that women experience during pregnancy, Gardner said. “These pathogens have the unusual ability to cross the placenta and directly infect the developing foetus, including the brain.”
While more common infections during pregnancy have been associated with autism and other neurodevelopmental conditions in simple comparisons of exposed and unexposed children, those associations largely disappear when studies compare siblings.
“Those findings suggest that the infection itself is probably not causing the increased risk,” Gardner said. “Instead, characteristics shared within families, including genetic factors that may influence both susceptibility to infection and neurodevelopment, may explain the association.”
With TORCH infections, however, the association remained when the research team compared siblings. “That provides much stronger evidence that the infection itself can contribute to autism and intellectual disability,” she added.
Strategies highlighting maternal TORCH infection risks may have a meaningful effect on the incidence of autism spectrum disorders (ASD), said Mark Schleiss, MD, of the University of Minnesota in Minneapolis, in the United States, in an accompany editorial.
In particular, awareness and knowledge of CMV’s risks “are disturbingly low”, Schleiss warned. Nearly half of the study’s TORCH infections were CMV (49.9%), with a rate of one to two cases per 10 000 births. Children with a congenital CMV infection were nearly five times more likely than children without a TORCH infection to develop autism (HR 4.57).
“CMV should receive higher public health emphasis as a preventable cause of intellectual disability and ASDs, and a major step forward would be realised by passage of the bipartisan Stop CMV Act of 2025 in the US,” Schleiss said. While a congenital CMV vaccine might prevent only a subset of ASD cases, he noted, it could be considered as a vaccine to prevent autism.
Expanding beyond TORCH infections to target other pathogens that endanger foetal brain development could lead to new vaccines “that may, by improving neurodevelopmental outcomes, help future generations of children achieve their full potential”, Schleiss added.
Gardner’s team analysed health records for 3 666 002 children born between 1987 and 2021, 975 of whom were diagnosed with congenital TORCH infections in the first year of life, for a rate of 2.5 per 10 000 births. Among those with TORCH infections, 54.6% were male, compared with 51.4% of those without infections.
There were 5.39 autism diagnoses per 1 000 person-years among those with TORCH infections, compared with 1.71 per 1 000 person-years among those without TORCH infections. Intellectual disability diagnosis rates were 8.22 per 1 000 person-years with TORCH infection and 0.68 without infection.
After a congenital TORCH infection, autism with intellectual disability was more likely than autism without intellectual disability (HR 6.23). Even when a TORCH infection wasn’t linked with either autism or intellectual disability, the researchers found lasting cognitive performance deficits among adolescents. Among 420 such kids, a TORCH infection was linked to a drop of -1.50 points in their school grades.
The study showed a surprising lack of association between TORCH infections and neuropsychiatric outcomes, including non-affective psychosis such as schizophrenia, Gardner noted. A slightly increased risk of attention deficit-hyperactivity disorder disappeared in the study’s sibling comparison analysis, while obsessive compulsive disorder showed no links to TORCH infections.
“That specificity is important, because it suggests that TORCH infections do not simply increase the risk of neurodevelopmental conditions in general,” Gardner said. “Rather, their effects may be more closely tied to particular aspects of brain development and cognitive function.”
Study limitations include a lack of newborn screening for TORCH infections, which prevented outcomes assessment in those with asymptomatic infections. In addition, inability to establish timing of infections relative to factors such as birth complications may have weakened some associations.
Study details
Congenital TORCH Infections and Neurodevelopmental Outcomes
Hugo Sjöqvist, Christina Dalman, David Mataix-Cols et al.
Published in JAMA Paediatrics on 21 September 2026
Abstract
Importance
Congenital TORCH (toxoplasmosis, syphilis, rubella, cytomegalovirus, or herpes simplex) infections are established causes of severe foetal injury, yet their population-level contribution to neurodevelopmental and psychiatric outcomes, independent of familial confounding, remains yet to be quantified.
Objective
To investigate whether congenital TORCH infections are associated with neurodevelopmental and psychiatric outcomes, including autism and intellectual disability, as well as academic performance, using population-based and sibling-controlled analyses.
Design, Setting, and Participants
This nationwide, population-based cohort study with sibling comparisons was conducted using linked data from Swedish national health, birth, insurance, education, and death registers. Individuals born in Sweden between 1987 and 2021, including individuals diagnosed with a congenital TORCH infection, were included. Full siblings were identified for within-family comparisons. Individuals were followed up from birth until death, emigration, or December 31, 2023. Data were analysed from November 2025 to July 2026.
Exposure
Clinically diagnosed congenital TORCH infections, identified from national registers, including cytomegalovirus, Toxoplasma gondii, rubella, and herpes simplex viruses.
Main Outcomes and Measures
Autism (with and without co-occurring intellectual disability), intellectual disability (by severity), attention-deficit/hyperactivity disorder, obsessive-compulsive disorder, Tourette and chronic tic disorders, non-affective psychosis, and standardized school grades at age 16 years were assessed. Associations were estimated with hazard ratios (HRs) using Cox regression and sibling-comparison models.
Results
Among 3 666 002 individuals born in Sweden, including 3 665 027 individuals without TORCH infections (1 883 911 male [51.4%]; mean [SD] follow-up, 20.50 [11.15] years) and 975 individuals with TORCH infections (532 male [54.6%]; mean [SD] follow-up, 18.70 [10.39] years), congenital TORCH infections were associated with increased risks of intellectual disability (HR, 7.22; 95% CI, 6.14-8.49) and autism (HR, 3.10; 95% CI, 2.55-3.76), with similar or greater HRs in sibling comparisons (intellectual disability: HR, 11.28; 95% CI, 5.97-21.33; autism: HR, 3.19; 95% CI, 1.97-5.15). Risks increased with greater severity of intellectual disability, with HRs ranging from 3.01; 95% CI, 2.21-4.11 for mild to 23.51; 95% CI, 18.06-30.60 for severe to profound intellectual disability) and were greater for autism with (HR, 6.23; 95% CI, 4.69-8.28) vs without (HR, 1.97; 95% CI, 1.51-2.57) co-occurring intellectual disability. No consistent associations were observed for obsessive-compulsive disorder, and associations with attention-deficit/hyperactivity disorder attenuated in sibling analyses. Individuals who were exposed had lower school grades in adolescence (TORCH-associated decrease = −1.50 points; 95% CI, −2.60 to −0.40 points).
Conclusions and Relevance
In this study, congenital TORCH infections were rare but associated with intellectual disability and autism, with evidence of broader associated cognitive outcomes extending beyond diagnosed conditions, underscoring the importance of preventing specific vertically transmitted infections.
JAMA Paediatrics article – Congenital TORCH Infections and Neurodevelopmental Outcomes (Open access)
See more from MedicalBrief archives:
Early autism signs can detected by routine prenatal ultrasound – Israeli study
Should there be universal screening for the often ignored CMV in newborns?
Genital herpes in pregnancy linked with double the risk of autism