Highlight
In this prospective cohort of 642 children with confirmed congenital CMV infection, adverse long-term outcomes after maternal primary infection were observed only when infection occurred in the first trimester.
Autism spectrum disorder was diagnosed in 11 children (1.7%), all symptomatic at birth; all cases after maternal primary infection followed first-trimester exposure.
Temporal lobe white matter abnormalities, particularly involving the temporal poles, were consistently present on postnatal MRI in children with autism spectrum disorder and were already detectable prenatally in those who underwent fetal MRI.
The findings support a narrow developmental window of fetal vulnerability and suggest a potential role for neuroimaging in prenatal and neonatal risk stratification.
Background
Congenital cytomegalovirus infection remains the most common congenital infection worldwide and the leading non-genetic cause of sensorineural hearing loss in childhood. Its clinical spectrum is broad, ranging from asymptomatic infection to severe multisystem disease, microcephaly, neurodevelopmental delay, epilepsy, cerebral palsy, and permanent hearing impairment. For obstetricians, pediatricians, infectious disease physicians, and fetal medicine specialists, the major unresolved challenge has been risk stratification: which fetuses are most likely to sustain clinically meaningful injury, and at what gestational stage is the fetal brain most vulnerable?
The timing question is biologically plausible and clinically important. The first trimester is a period of rapid neurogenesis, neuronal migration, and early organization of cortical and subcortical structures. Viral injury during this period could plausibly produce more severe and lasting effects than infection later in pregnancy. Yet in routine counseling, the precision of trimester-specific risk estimates has often been limited by retrospective designs, heterogeneous maternal serology, incomplete follow-up, and inconsistent neuroimaging.
A possible association between congenital CMV and autism spectrum disorder has also been debated for decades. Prior literature largely consisted of case reports, small series, and retrospective analyses, leaving uncertainty about whether CMV is a causal contributor, a marker of broader neurodevelopmental disruption, or simply an incidental co-occurrence in a small subset of affected children. This study by Leruez-Ville and colleagues addresses these gaps using a large, prospective, single-center cohort with centralized maternal serological dating and standardized long-term follow-up.
Study Design
Design and setting
This was a prospective observational cohort study conducted at a single tertiary referral center in France between 2001 and 2024. The investigators included children with confirmed congenital CMV infection who underwent standardized follow-up to 48 months of age.
Population
The full cohort included 642 children with confirmed congenital CMV infection. Maternal CMV infections were centrally classified and dated using serological assessment, an important methodological strength because accurate timing of maternal primary infection is essential when interpreting trimester-specific fetal risk.
Among the 642 children, 504 (78.5%) were exposed to maternal primary infection with known timing. Within this timed primary-infection subgroup, 288 children (57.1%) had first-trimester exposure, 144 (28.6%) second-trimester exposure, and 72 (14.3%) third-trimester exposure.
Assessments
Children underwent standardized longitudinal evaluations that included audiological, neurological, behavioral, and neuroimaging assessments. Fetal brain MRI was available in antenatally diagnosed cases, allowing the investigators to examine prenatal imaging correlates of later neurodevelopmental outcomes.
Outcomes
The main outcomes of interest were hearing loss, neurodevelopmental impairment, and autism spectrum disorder. The study also examined neuroanatomical correlates, especially whether specific imaging patterns were associated with autism spectrum disorder.
Key Findings
Trimester of maternal infection strongly shaped long-term outcome
The central message of the study is striking: long-term congenital CMV-related sequelae after maternal primary infection, including hearing loss and neurodevelopmental impairment, were observed exclusively in children exposed during the first trimester. No such long-term adverse outcomes were identified after second- or third-trimester primary infection in this cohort.
This finding suggests that gestational timing is not merely one risk modifier among many, but may define a critical window during which CMV can inflict durable fetal brain injury. From a counseling perspective, this is highly consequential. It supports a more nuanced and biologically grounded discussion with pregnant patients who acquire primary CMV infection, especially early in gestation.
Autism spectrum disorder signal
Autism spectrum disorder was diagnosed in 11 of 642 children, corresponding to a prevalence of 1.7%. All 11 children were symptomatic at birth, indicating that autism spectrum disorder in this cohort did not arise from otherwise silent congenital infection. Among cases linked to maternal primary infection, all occurred after first-trimester exposure.
When compared with estimates from the general French population, the prevalence of autism spectrum disorder among children with congenital CMV was approximately fourfold higher, with an odds ratio of 4.25 and a 95% confidence interval of 1.63 to 21.33. The point estimate suggests a clinically meaningful association, although the wide confidence interval indicates imprecision due to the limited number of events.
These data do not imply that congenital CMV is a common cause of autism spectrum disorder at the population level. Rather, they suggest that among children with congenital CMV, particularly those symptomatic at birth after first-trimester exposure, autism spectrum disorder may represent part of the severe neurodevelopmental phenotype.
Neuroimaging correlates: temporal lobe white matter abnormalities
Perhaps the most intriguing finding is the imaging consistency. In all children diagnosed with autism spectrum disorder, postnatal MRI demonstrated temporal lobe white matter abnormalities, especially involving the temporal poles. In fetuses who had undergone prenatal MRI, these abnormalities were already present antenatally.
The reverse observation is also important. When investigators reviewed all fetal MRIs in the prenatally diagnosed subgroup, temporal lobe white matter abnormalities were present in only a minority of cases, and no child without temporal lobe abnormality developed autism spectrum disorder. This pattern does not prove that the imaging lesion is a deterministic biomarker, but it does suggest that temporal lobe white matter injury may be a useful risk-enrichment feature.
The temporal lobes are plausible substrates for social communication and language-related outcomes. White matter injury in these regions could disrupt long-range connectivity, auditory-linguistic processing, and socioemotional network development, all of which have been implicated in autism spectrum disorder. The consistency between prenatal and postnatal imaging also strengthens the argument that these lesions reflect true fetal injury rather than postnatal evolution alone.
Clinical phenotype of affected children
All children with autism spectrum disorder were symptomatic at birth. Although the abstract does not provide a full breakdown of neonatal manifestations, this observation suggests that autism spectrum disorder emerged in the context of a broader clinically evident congenital CMV syndrome rather than in mildly affected or asymptomatic infants. Clinicians should therefore be cautious not to overgeneralize the autism finding to all infants with congenital CMV.
The same principle likely applies to hearing loss and broader neurodevelopmental impairment. The study supports a concentrated burden of sequelae in first-trimester infection, but the individual risk within that subgroup likely remains heterogeneous and is influenced by fetal imaging findings, symptom severity at birth, and perhaps other host and viral factors not fully captured in the abstract.
Clinical Interpretation
Why the first trimester matters
The findings align with developmental neurobiology. Early gestation is marked by neural progenitor proliferation, regional patterning, early cortical organization, and the establishment of axonal pathways. CMV is known to infect neural progenitor cells and to alter cell cycling, differentiation, and migration. Injury during this period is therefore more likely to produce structural and connectivity abnormalities than infection later in pregnancy, when some foundational developmental processes are further advanced.
This also helps explain why second- and third-trimester infection may still result in congenital infection but not necessarily the same burden of permanent neurological sequelae. Infection later in gestation may affect hearing or development in some series, but this study suggests that, at least after well-dated maternal primary infection in this cohort, the most consequential long-term injuries were confined to first-trimester exposure.
Implications for prenatal counseling
For maternal-fetal medicine specialists, the study offers a more refined framework for counseling after primary CMV infection. First-trimester maternal primary infection should be recognized as the highest-risk scenario for significant fetal neurological sequelae. In contrast, when primary infection clearly occurs in the second or third trimester, this cohort suggests a substantially lower risk of major long-term neurodevelopmental complications.
That said, counseling should remain individualized. A single-center observational cohort, even a large one, cannot eliminate all uncertainty. Moreover, the study does not justify complacency when fetal ultrasound or MRI is abnormal, regardless of gestational timing. Imaging findings still matter, and direct fetal assessment remains central to management.
Implications for neonatal follow-up
The data support intensive developmental surveillance for infants with congenital CMV who were exposed in the first trimester, especially those symptomatic at birth and those with temporal lobe white matter abnormalities on prenatal or postnatal MRI. Follow-up in such children should extend beyond hearing assessments to include early neurodevelopmental and behavioral screening, with explicit attention to social communication, language trajectory, and early signs of autism spectrum disorder.
Earlier recognition has practical value. Children identified as being at elevated developmental risk can be referred sooner for audiology, speech-language evaluation, developmental pediatrics, early intervention services, and autism-specific assessment when indicated.
Strengths and Limitations
Major strengths
This study has several important strengths. First, the cohort is large for congenital CMV research, particularly for analyses focused on timing of maternal infection. Second, centralized serological dating strengthens exposure classification. Third, prospective, standardized follow-up to 48 months reduces recall bias and improves outcome ascertainment. Fourth, the inclusion of fetal and postnatal MRI allows a meaningful clinicoradiologic correlation that is often missing from prior literature.
Key limitations
Several limitations should temper interpretation. The study was conducted at a single tertiary referral center, which may introduce referral bias and limit generalizability. Tertiary cohorts often overrepresent more severe or prenatally recognized cases. The number of autism spectrum disorder cases was small, reflected in the wide confidence interval around the odds ratio. Therefore, the magnitude of the autism association should be interpreted cautiously.
Autism diagnosis at up to 48 months is clinically relevant, but some neurodevelopmental phenotypes evolve beyond early childhood. Longer follow-up may reveal additional cognitive, behavioral, executive, or psychiatric outcomes not captured by age 4 years. The abstract also does not detail antiviral treatment exposure, socioeconomic factors, coexisting prematurity or neonatal complications, or the exact diagnostic methods used for autism spectrum disorder, all of which could influence observed outcomes.
Finally, observational data cannot by themselves establish causality. The imaging-pathology-behavior relationship is compelling, but whether temporal lobe white matter abnormalities are a direct mechanistic mediator of autism spectrum disorder, a marker of more diffuse brain injury, or both remains unresolved.
Relationship to Existing Literature
Prior work has firmly established congenital CMV as a major cause of sensorineural hearing loss and neurodevelopmental morbidity. Reviews and guideline statements from pediatric and infectious disease societies have emphasized the heterogeneity of outcome and the need for ongoing audiologic monitoring even in some asymptomatic infants. This study adds granularity by suggesting that, after maternal primary infection, the first trimester is the dominant period for severe long-term sequelae.
The autism signal is also noteworthy in the context of earlier, less definitive reports. Previous case reports and retrospective studies proposed a link but were often limited by selection bias, incomplete maternal dating, and lack of imaging correlation. Leruez-Ville and colleagues advance the field by tying autism spectrum disorder to a specific exposure window and a reproducible neuroanatomical pattern.
These results are also consistent with a broader shift in congenital infection research toward combining precise maternal infection dating, fetal imaging, and structured long-term developmental follow-up. Such integrated datasets are more informative for counseling than crude overall transmission or symptom rates alone.
Practice Takeaways
For clinicians, several messages are immediately actionable. First, accurate dating of maternal CMV primary infection matters and should be pursued whenever possible using expert serological interpretation. Second, first-trimester primary infection warrants particularly careful fetal evaluation and longitudinal counseling. Third, fetal or neonatal MRI findings in the temporal lobes may help identify a subgroup at heightened risk for autism spectrum disorder and other neurodevelopmental consequences. Fourth, infants with symptomatic congenital CMV should receive structured long-term developmental surveillance, not hearing follow-up alone.
At the same time, clinicians should avoid overstatement. Most children with congenital CMV do not develop autism spectrum disorder. Even within first-trimester exposure, risk is not uniform. The study supports risk stratification, not deterministic prediction.
Funding and ClinicalTrials.gov
The abstract does not report funding details or a ClinicalTrials.gov registration number. Readers should consult the full article for complete disclosure regarding funding, ethics approval, and any protocol registration.
Conclusion
This prospective cohort provides important evidence that the most serious long-term sequelae of congenital CMV after maternal primary infection are concentrated in first-trimester exposure. Hearing loss, neurodevelopmental impairment, and all autism spectrum disorder cases linked to primary maternal infection occurred in this early gestational window. The consistent association between autism spectrum disorder and temporal lobe white matter abnormalities, visible both prenatally and postnatally, is particularly provocative and clinically useful.
For obstetric and pediatric practice, the study sharpens risk counseling, reinforces the value of expert fetal imaging, and argues for targeted developmental surveillance of high-risk infants. For research, it highlights the need for multicenter validation, longer-term neurobehavioral follow-up, and mechanistic studies clarifying how early CMV-related temporal lobe injury shapes later developmental trajectories.
References
1. Leruez-Ville M, Grevent D, Bourgon N, Chatzakis C, Parodi M, Fourgeaud J, Veyrenche N, Bahi-Buisson N, Pichon C, Magny JF, Boddaert N, Ville Y. Maternal cytomegalovirus infection in the first trimester of pregnancy: timing, fetal brain injury, and long-term neurodevelopmental outcomes including autism spectrum disorder. American Journal of Obstetrics and Gynecology. 2026-05-04. PMID: 42092723.
2. Rawlinson WD, Boppana SB, Fowler KB, Kimberlin DW, Lazzarotto T, Alain S, Daly K, Doutré S, Gibson L, Giles ML, et al. Congenital cytomegalovirus infection in pregnancy and the neonate: consensus recommendations for prevention, diagnosis, and therapy. Lancet Infectious Diseases. 2017;17(6):e177-e188.
3. Kimberlin DW, Lin CY, Sánchez PJ, Demmler GJ, Dankner W, Shelton M, Jacobs RF, Vaudry W, Pass RF, Kiell JM, et al. Effect of ganciclovir therapy on hearing in symptomatic congenital cytomegalovirus disease involving the central nervous system: a randomized, controlled trial. Journal of Pediatrics. 2003;143(1):16-25.
4. Kimberlin DW, Jester PM, Sánchez PJ, Ahmed A, Arav-Boger R, Michaels MG, Ashouri N, Englund JA, Estrada B, Jacobs RF, et al. Valganciclovir for symptomatic congenital cytomegalovirus disease. New England Journal of Medicine. 2015;372(10):933-943.
5. Dollard SC, Grosse SD, Ross DS. New estimates of the prevalence of neurological and sensory sequelae and mortality associated with congenital cytomegalovirus infection. Reviews in Medical Virology. 2007;17(5):355-363.
