Congenital Cytomegalovirus Infection

Transplacental fetal infection by human cytomegalovirus, the main non-genetic cause of childhood sensorineural hearing loss and a phenocopy of monogenic disorders with intracranial calcifications, microcephaly and leukoencephalopathy.

Literature and regulatory information updated: 17 October 2026.

0. Initial Reference Table

The following table brings together the genetic phenocopies that the assessment of suspected cCMV requires to be considered when the infection is not confirmed, developed in section 6: hereditary primary microcephaly, leukodystrophies, Aicardi-Goutières syndrome and type I interferonopathies, and non-syndromic genetic deafness.

1. Introduction

1.1. The place of congenital CMV infection in a medical genetics handbook

Congenital cytomegalovirus infection (cCMV) is diagnosed and treated mainly by fetal medicine, neonatology, paediatric infectious diseases and virology, but the geneticist is often involved before the diagnosis has been established, because the phenotype of the affected fetus or infant overlaps with that of several genetic disorders [23]. Three situations make this article necessary: cCMV is the most common severe intrauterine infection and the most common acquired cause of prelingual deafness in high-income countries [5, 23]; the intracranial calcifications, polymicrogyria, microcephaly and leukoencephalopathy it produces have autosomal recessive monogenic phenocopies with a 25% recurrence risk [4, 12]; and a genetic aetiology of deafness and cCMV may coexist in the same child, so a diagnosis of cCMV does not obviate genetic assessment [3]. The 2017 international consensus described cCMV as the most frequent, and underappreciated, infectious cause of malformation in the newborn in developed countries [24].

1.2. Historical milestones

The propagation in tissue culture of a cytopathogenic virus obtained from so-called human salivary gland virus disease was published by Margaret Smith in 1956, which corresponds to the isolation of human CMV [25]. In 1962, Weller and Hanshaw published the virological and clinical observations of cytomegalic inclusion disease, the classic name for the symptomatic neonatal form [2]. The sources consulted for this article record only the metadata of these two founding articles, so the details of the original observations are not reproduced. Subsequent developments included the replacement of viral culture of urine, the former reference standard, by PCR on urine or saliva in the first 3 weeks of life [3, 23], the demonstration in randomised trials of the Collaborative Antiviral Study Group (CASG) of the benefit of intravenous ganciclovir for 6 weeks (2003) and oral valganciclovir for 6 months (2015) in NB with disease [9, 26], and the first randomised placebo-controlled trial of valaciclovir to prevent vertical transmission after early maternal primary infection (2020) [27].

1.3. Nomenclature and operational definitions

Case definitions have changed in recent years. The 2026 AAP clinical report replaces "symptomatic cCMV" with cCMV disease, which denotes signs or sequelae attributable to the infection on physical examination or on complementary investigations (imaging, laboratory, audiology), and "asymptomatic cCMV" with cCMV infection; it adds the category of cCMV with isolated SNHL, when hearing loss is the only sign, and that of cCMV disease with SNHL, when hearing loss coexists with other sequelae [3]. The 2017 European expert consensus proposed distinguishing severe, moderate and mild disease and separating NB identified by clinical suspicion from those identified by screening, because the two populations differ in severity [28]. The AAP also adopts its own auditory terminology: deaf for thresholds ≥91 dB in both ears, hard of hearing for 21–90 dB and unilateral deafness for ≥91 dB in one ear with ≤20 dB in the other [3].

Maternal primary infection is the first infection in a seronegative woman; non-primary infection comprises reactivation of latent virus and reinfection by a new strain [3]. To date primary infection in relation to pregnancy, the literature uses the preconceptional (2–8 weeks before conception in the 2020 meta-analysis), periconceptional (about 2 weeks before and after conception), 1st-, 2nd- and 3rd-trimester periods [8, 29]. The term "pseudo-TORCH" has historically been applied to different entities: the Baraitser-Reardon microcephaly–intracranial calcification syndrome, now considered in most cases to be Aicardi-Goutières syndrome (AGS), band-like calcification with polymicrogyria due to OCLN variants (pseudo-TORCH type 1) and the interferonopathies due to USP18 deficiency (type 2) and STAT2 gain of function (type 3) [4, 11, 13, 14].

1.4. Scope of the article

This article covers epidemiology, transmission, the clinical picture with emphasis on SNHL, maternal, fetal and neonatal diagnosis, universal and targeted newborn screening strategies, secondary prevention with valaciclovir in pregnancy, neonatal treatment with valganciclovir, trials with letermovir and vaccines under study. The differential diagnosis includes the genetic phenocopies (AGS1–AGS9, including LSM11 and RNU7-1, USP18, OCLN, STAT2, ISG15, JAM3, RNASET2, COL4A1/COL4A2, Coats plus and SNORD118, NDE1, ANKLE2) and the genetic causes of deafness (GJB2/GJB6, mitochondrial DNA, SLC26A4).

2. Molecular Genetics and Pathophysiology

2.1. The virus, latency and horizontal transmission

Human CMV is a ubiquitous betaherpesvirus, specific to the human species, without seasonality, that establishes lifelong latency after primary infection; subclinical reactivation is frequent and reinfection by new strains is possible, so natural immunity does not prevent later infections [3]. Horizontal transmission occurs through close contact with saliva, urine, breast milk and genital secretions, sexual activity being a risk factor; transmission through blood, transfusion or transplantation is rare [3, 23]. Young children shed the virus at high concentrations and for prolonged periods: in a French study from 2009–2012, about 40% (95% CI 34.6–46.1) of children in day care were shedding CMV, and transmission from child to parents was confirmed by viral isolation, serology and molecular epidemiology [3]. In the USA, seroprevalence between 1 and 5 years was 29% (95% CI 24.4–33.9) in 2017–2020, and the Centers for Disease Control and Prevention (CDC) estimate that about 1 in 3 children is infected by the age of 5 years, keeping the virus in saliva and urine for months [3, 30]. Breast milk transmits clinically silent infection to the healthy term NB, but infection acquired at delivery or through breastfeeding makes it difficult, after the first weeks, to distinguish congenital from postnatal infection [3, 23].

2.2. Vertical transmission and maternal immunity

Transmission to the fetus occurs transplacentally after maternal primary infection or after non-primary infection [3]. Preconceptional immunity reduces the risk of cCMV: in a cohort of 3461 multiparous women from a population with a high cCMV rate (1993–1998), cCMV occurred in 18/604 (3.0%) children of initially seronegative mothers and in 29/2857 (1.0%) children of mothers who were already immune, with an adjusted relative risk of 0.31 (95% CI 0.17–0.58), which corresponds to a 69% reduction [31]. Immunity protects against intrauterine transmission, but not against disease once the fetus is infected: recent studies and meta-analyses show similar hearing and neurological outcomes in infants infected after primary or non-primary infection, contrary to what older studies suggested [3, 8]. In primary infection, the transmission rate increases with gestational age, while the severity of fetal lesions decreases; this inverse relationship, quantified in section 3, resolves the apparent contradiction of older texts that describe transmission as independent of gestational age [3, 23, 29].

2.3. Pathogenesis of fetal lesions

The chronology of brain development explains the relationship between the timing of infection and the type of lesion. Neuronal proliferation takes place between 8 and 16 weeks, migration to the cortex between 12 and 20 weeks and cortical organisation from 22 weeks; astrocytes predominate at 26 weeks and differentiate into oligodendrocytes in the first half of the 3rd trimester, and the germinal zone is exhausted at 32–34 weeks [32]. An early infection reduces neurons and glia, whereas a lesion after 24–26 weeks no longer affects migration; thus, 1st-trimester infection is associated with lissencephaly, polymicrogyria, cerebellar hypoplasia and calcifications, and 3rd-trimester infection with white matter inflammation and periventricular or subependymal cysts, and persistent infection may combine the two patterns [32]. The 2024 European consensus classifies neuroimaging abnormalities into two types: inflammatory or destructive changes, resulting from the direct effect of the virus or the inflammatory response (lenticulostriate vasculopathy, caudothalamic, temporal or frontal germinolytic pseudocysts, septations of the occipital horns, ventriculomegaly, periventricular calcifications, T2 white matter hyperintensity), and disorders of brain development (cortical malformations, typically polymicrogyria or deficient sulcation, and cerebellar hypoplasia) [8].

The pathophysiology of SNHL is not clear; the most likely explanation is infection and degradation of the sensory structures of the inner ear, which feeds the debate on whether isolated SNHL should be considered a central nervous system (CNS) manifestation equivalent to that of the NB included in the randomised trials [28]. Vestibular involvement, present in about 17% of children with cCMV, may be congenital or late-onset and may fluctuate, improve or worsen [8]. The viral load in neonatal blood correlates with symptomatic disease and with sequelae, especially SNHL, but without a consensus threshold (section 11) [8].

2.4. Molecular basis of the genetic phenocopies

The similarity between cCMV and several monogenic disorders has a common mechanistic explanation: the type I interferonopathies induce type I interferon and interferon-stimulated genes as if viral DNA or RNA were present, reproducing the host response to an infection [33]. In AGS, the genes TREX1, RNASEH2A, RNASEH2B, RNASEH2C, SAMHD1, ADAR and IFIH1 take part in the metabolism or sensing of self nucleic acids [4, 33]. Biallelic variants in LSM11 and RNU7-1, components of the replication-dependent histone pre-mRNA processing complex, alter the processing of canonical histones and the distribution of nuclear cGAS and increase interferon signalling through the cGAS-STING pathway (AGS8 and AGS9) [16]. USP18 is an interferon-induced protease that removes ISG15 and is the main brake on type I interferon signalling: recruited by STAT2 to the IFNAR2 receptor, it blocks access of JAK1, and its absence produces unmitigated inflammation, lethal in the perinatal period [12, 34]. The homozygous STAT2 p.(Arg148Trp) variant makes cells hypersensitive to interferon α/β through failure of interaction with USP18, a gain of function with recessive expression [35]. Intracellular ISG15 stabilises USP18; its absence explains, in inherited ISG15 deficiency, the coexistence of interferonopathy with intracranial calcifications, susceptibility to mycobacteria and necrotising skin lesions [12, 36]. A 2026 preprint, without peer review, proposes heterozygous loss-of-function variants in PTPN1, a negative regulator of the interferon pathway, as a cause of late-onset AGS with reduced penetrance [37].

A second group of phenocopies results from injury to the blood–brain barrier or small vessels. Occludin, encoded by OCLN, is a tight junction protein, and the neuropathology of band-like calcification with polymicrogyria shows calcification mainly associated with vessels [38]. JAM-C, encoded by JAM3, is an endothelial tight junction protein whose loss causes recurrent cerebral haemorrhages from fetal life [12, 39]. COL4A1 and COL4A2 cause genetic small vessel disease of the brain, eyes and kidneys, with antenatal, perinatal or postnatal haemorrhage and ischaemic injury [19]. RNASET2 leukoencephalopathy produces anterior temporal cysts and multifocal white matter lesions essentially indistinguishable from the brain injury of cCMV [12, 19].

2.5. Recurrent variants in the differential diagnosis: standardised nomenclature

The table converts the historical nomenclature of the variants most often cited in the differential diagnosis into HGVS, with MANE Select transcript and version; the conversions were made in VariantValidator (GRCh38) and, for mitochondrial DNA, in ClinVar, and not by manual calculation [40, 41]. The pathogenicity classification of the nuclear variants was not reassessed for this article and follows the respective GeneReviews.

2.6. Coexistence of infection and genetic aetiology

A positive CMV PCR in a NB with SNHL does not exclude a concurrent genetic cause, because the two may coexist; the AAP stresses that a diagnosis of cCMV does not obviate genetic assessment [3]. The GeneReviews chapter on GJB2 deafness includes the result of CMV testing, when performed, in the initial assessment [20], and the Ontario universal screening programme tests the same dried blood spot for CMV and for the most common genetic causes of deafness [3]. The Food and Drug Administration (FDA) approval, on 23 April 2026, of a gene therapy for deafness due to biallelic OTOF variants, which requires an ear without a previous cochlear implant, gives direct therapeutic consequences to the identification of a coexisting genetic cause [5, 42]. The sources consulted do not describe host variants with established value as modifiers of the risk of transmission or of the severity of cCMV.

3. Clinical Spectrum and Manifestations by System

3.1. Maternal infection

Primary infection is asymptomatic in most women; in immunocompetent children and adults the infection is, as a rule, subclinical [23, 43]. Primary infection occurs in about 1–2% of pregnancies in the USA and Europe, and in 1–4% of seronegative pregnant women according to the Society for Maternal-Fetal Medicine (SMFM) [3, 8, 43]. In non-primary infection, CMV shedding was detected in 21.5% (95% CI 12.7–30.3) of seropositive pregnant women in a meta-analysis, with vertical transmission probably below 3.5% [8]. The meta-analysis by Kenneson and Cannon (2007) estimated transmission of 32% after primary infection and 1.4% after recurrent infection [6].

3.2. Transmission and fetal injury according to the gestational age of primary infection

The meta-analysis by Chatzakis and colleagues (2020; 10 studies, 2942 fetuses for transmission) quantified the inverse relationship between transmission and severity. Vertical transmission was 5.5% (95% CI 0.1–10.8) for preconceptional infection, 21.0% (8.4–33.6) for periconceptional, 36.8% (31.9–41.6) in the 1st trimester, 40.3% (35.5–45.1) in the 2nd and 66.2% (58.2–74.1) in the 3rd [29]. When transmission occurred (10 studies, 796 fetuses), fetal injury, defined by prenatal CNS findings that led to termination of pregnancy for medical reasons (TOP) or by neurological symptoms at birth, occurred in 28.8% (2.4–55.1) for periconceptional infection, 19.3% (12.2–26.4) in the 1st trimester, 0.9% (0–2.4) in the 2nd and 0.4% (0–1.5) in the 3rd; SNHL occurred in 22.8% (15.4–30.2), 0.1% (0–0.8) and 0% (0–0.1) after infection in the 1st, 2nd and 3rd trimesters [29]. The 2024 European consensus cites the confidence interval of 0–2.1% for the 3rd trimester, a divergence resolved in favour of the original abstract [8, 29]. For long-term sequelae after primary infection, the same consensus gives 23% (95% CI 15.4–30.2) in the 1st trimester, 0.1% in the 2nd and 0% in the 3rd [8].

The German prospective cohort of Enders and colleagues (2011; 248 pregnancies with primary infection dated by seroconversion, avidity or symptoms), reproduced in Table 1 of the AAP report, shows the same relationship in absolute numbers [3, 44].

The original abstract gives 34.5% (10/29) for periconception, a value that the AAP table rounds to 34.3% [3, 44]. In this cohort, 10.3% of infected fetuses or NB had severe manifestations and 22.8% any sign, and none of those infected after preconceptional or 3rd-trimester primary infection had symptoms at birth [44]. The French prospective cohort of Faure-Bardon and colleagues (2011–2017; 255 women with primary infection, 260 infected fetuses or NB, prospective dating in 86%, structured follow-up up to 48 months, median 24) confirmed that sequelae are confined to 1st-trimester infection [3, 45].

The 15 neurological sequelae of the 1st trimester were 11 vestibular disorders, 1 severe global delay, 1 monoplegia of the arm, 1 severe autistic presentation and 1 spastic diplegia; 11, 9 and 3 children had audiology that could not be interpreted because of otitis media [3]. The absence of sequelae in the 2nd and 3rd trimesters in a cohort of this size does not demonstrate zero risk: a retrospective study recorded late-onset SNHL in 3/140 (2.1%) children after 2nd-trimester primary infection [8]. The SMFM, based on earlier series, states that among the children of women with primary infection 18% are symptomatic at birth, and that up to 25% of those not symptomatic develop sequelae in the first 2 years; after recurrent infection, less than 1% are symptomatic, 8% have sequelae at 2 years and 14% at 5 years [43].

3.3. Fetal manifestations

Ultrasound suggests infection in fewer than half of infected fetuses; in a cohort of 600 women with primary infection, 8.5% of fetuses had ultrasound abnormalities, a proportion that rose to 14.9% after review of the scans of NB with confirmed infection, with a positive predictive value of ultrasound of 35% [43]. The table reproduces the frequency of ultrasound abnormalities in confirmed cases of cCMV compiled by the SMFM [43].

Conversely, hyperechogenic bowel, present in about 0.8% of pregnancies in the 2nd and 3rd trimesters, is associated with cCMV in about 2–3% of cases; in the French multicentre study of 682 fetuses with this finding, 2.8% had a viral infection (15 due to CMV and 4 due to parvovirus), 3% cystic fibrosis and 3.5% a significant chromosomal anomaly [46]. In an Italian series of 77 infected fetuses without major malformations, mild early lesions on fetal magnetic resonance imaging (MRI) appeared in 11.3% of those not treated [47].

3.4. Neonatal manifestations

Only 10–15% of NB with cCMV have recognisable findings in the neonatal period; in a meta-analysis of 2098 infants, disease occurred in 10.8% (95% CI 7.40–15.85) [3]. The most frequent findings are petechiae, hepatosplenomegaly, microcephaly, small for gestational age (SGA) weight and SNHL; in the CNS, chorioretinitis, microcephaly and seizures; and, outside the CNS, hepatitis, cholestasis with direct hyperbilirubinaemia, thrombocytopenia, purpura, gastrointestinal disease (strictures, perforation, diarrhoea, colitis) and, less often, pneumonitis [3]. The "blueberry muffin" rash, an expression of dermal haematopoiesis, is part of the list of signs in the European consensus [8]. Mortality in cCMV disease ranges between 0.5 and 12% depending on whether in-hospital, neonatal or infant mortality is measured [3]; the older estimate of about 20% perinatal deaths in NB with the typical picture comes from series predating antiviral treatment [23].

The European consensus defines the criteria for each sign: growth restriction and microcephaly below −2 standard deviations for gestational age, thrombocytopenia <100,000/µL, isolated leucopenia or neutropenia <1000/µL, aminotransferases ≥2 times the upper limit of normal, direct bilirubin >2 mg/dL, SNHL with a threshold >20 dB, unilateral or bilateral, and chorioretinitis or chorioretinal scar [8]. Microcephaly may be proportional to weight, in the context of symmetrical growth restriction, or relative (head circumference z-score minus weight z-score below −2), and only the latter has high specificity for CNS involvement and a poor neurological prognosis [8]. The classic series of Azam and colleagues (2001), reproduced in the Oxford Desk Reference, gives the frequency of findings in NB with symptomatic infection; the denominator is not given in the secondary source [23].

The mode of identification changes the phenotype. In the AAP series that compared NB with disease identified by screening (n=78) and those referred because of clinical suspicion (n=100), all signs were less frequent in the screened group, and 42% of them had a single finding [3].

3.5. Neuroimaging

The abnormalities described in cCMV include predominantly periventricular intracranial calcifications, periventricular and subependymal pseudocysts, polymicrogyria and other migration anomalies (lissencephaly), ventriculomegaly, temporal lobe abnormalities and cysts, cerebellar hypoplasia, lenticulostriate vasculopathy and white matter abnormalities [3, 23]. cCMV is the most common cause of the association between intracranial calcifications and polymicrogyria [19]. The white matter pattern was associated retrospectively with cCMV by PCR on Guthrie cards, but atypical cases exist [19]. Lenticulostriate vasculopathy is non-specific, also described in genetic anomalies, congenital toxoplasmosis, twin–twin transfusion and human immunodeficiency virus (HIV) infection [32].

3.6. Hearing

SNHL is the most common sequela of cCMV, and cCMV is the most common non-genetic cause of childhood SNHL [3, 5]. The AAP defines early SNHL as that present at birth or on the first diagnostic auditory brainstem response (ABR) up to 3 months, and late-onset SNHL as that appearing afterwards [3]. In NB with cCMV disease, SNHL occurs in 38.1% (95% CI 23.6–54.4) versus 4.7% (2.9–7.3) in NB without signs; the risk of hearing abnormalities throughout childhood is 40.7–44% and 7.4–11%, respectively, and systematic reviews record 27–74% and 7–11% at follow-up [3]. In a prospective cohort of children screened at birth, 12.7% had SNHL at birth and 4.5% developed late-onset SNHL; progression is frequent (>50%) and unilateral SNHL may become bilateral [8]. In the prospective cohort of infants without signs at birth, hearing abnormalities reached 7% at 3 months, 14% at 5 years and 25% at 18 years, but the incidence between 6 and 18 years did not differ from that of controls, and the risk of late-onset SNHL after 5 years in asymptomatic children does not differ from that of uninfected children [3, 8]. The AAP estimates that about 20% of children with cCMV will develop SNHL [3].

Older text sources give different figures because they measure different things: about 60% hearing deficit in survivors of symptomatic disease and 15% in infants without signs at birth in the 2017 Oxford Desk Reference, and deafness in about 50% of symptomatic and up to 10% of asymptomatic children in GeneReviews [5, 23]. SNHL associated with disease tends to be moderate to severe, isolated SNHL is more often unilateral, and late-onset SNHL is generally milder, asymmetrical and progresses more in the worse ear; sound-field behavioural audiometry may miss unilateral abnormalities, so the thresholds of each ear are monitored [3]. cCMV can cause an enlarged vestibular aqueduct and other malformations of the temporal bone, which limits the specificity of these findings for SLC26A4 deafness [22].

3.7. Vestibular function

Vestibular involvement affects about 17% of children with cCMV; early bilateral areflexia delays motor development, and the risk factors are maternal primary infection in the 1st trimester or of unknown timing, SNHL and periventricular cysts on MRI [8]. Vestibular dysfunction occurs in both severity groups and may affect head, postural and motor control [3].

3.8. Vision

In the cohort of Lanzieri and colleagues (2017), compiled by the AAP, 51% of the 76 NB with cCMV disease and ophthalmological data had some ocular abnormality, and 15% of the 66 with a known visual outcome were legally blind, concentrated in the subgroup with SNHL [3].

In the Ontario universal screening programme (2019–2022), 347 of 394 screen-positive NB underwent ophthalmological examination; only 2 (0.58% of positives; 3.9% of the 51 symptomatic) had attributable chorioretinal scars, both already identified because of systemic disease, and no asymptomatic NB had ocular findings [48].

3.9. Neurodevelopment, epilepsy and cerebral palsy

Lifelong disability (cognitive, motor or auditory) affects 40–58% of NB with cCMV disease and about 10–15% of NB with infection without signs [3, 7]. A systematic review found neurodevelopmental impairment in 30–66% of symptomatic children and performance equal to that of controls in asymptomatic children; one study suggests a higher incidence of autism spectrum disorder, attention deficit hyperactivity disorder and behavioural problems, especially after 1st-trimester infection [8]. Intellectual impairment (intelligence quotient, IQ, <70) and impaired adaptive behaviour occur in 29–43% of children with disease; epilepsy, in 1.5–37%, is more frequent in disease, and in a series of 19 children referred with disease 7 had epilepsy (mean age 20 months), all with cerebral palsy [3]. In the Australian cerebral palsy register, 9.6% (95% CI 6.8–13.3) of children had cCMV; in a Dutch national study with retrospective diagnosis on dried blood spots, about 1.5% of 5-year-old children with cCMV had cerebral palsy and 25% had impairment in at least one domain, versus 12% of children without CMV, with doubled healthcare costs [3]. Children with infection without signs up to 12–24 months did not differ from controls on developmental tests up to 18 years; those who had SNHL at 2 years had lower global IQ and receptive vocabulary, with intact non-verbal IQ, in a cohort predating newborn hearing screening [3].

4. Clinical Recognition and Red Flags

4.1. In pregnancy

Because maternal infection is as a rule asymptomatic, suspicion arises mainly from fetal ultrasound abnormalities [23, 43]. The findings that most often lead to investigation for CMV are hyperechogenic bowel, ventriculomegaly and brain calcifications and fetal growth restriction; hepatic calcifications, microcephaly and subependymal cysts are also described [43]. For isolated hyperechogenic bowel, the Oxford Desk Reference recommends offering cystic fibrosis carrier testing to the parents and maternal serology for congenital infection [46]. The European consensus does not recommend serology after 16 weeks, unless there are compatible ultrasound findings [8]. An occupation with exposure to young children, especially in day care, is a recognised risk factor [3, 8].

4.2. In the newborn

The 2024 European consensus recommends testing all children of mothers with suspected or confirmed primary infection in pregnancy and any NB with suspected hearing loss at birth (grade A), NB with fetal imaging abnormalities potentially associated with CMV, with symmetrical IUGR of unknown cause, very preterm (<32 weeks) or weighing <1500 g, and infants with unexplained compatible symptoms or laboratory or imaging abnormalities (grade D) [8]. The signs that should prompt testing are petechiae, hepatosplenomegaly, jaundice, microcephaly and thrombocytopenia, and the evidence is strongest for unilateral or bilateral SNHL; routine testing of all NB with IUGR or prematurity has a low yield [8]. The 2017 consensus did not recommend routine testing of SGA NB and noted that testing very preterm infants (<28 weeks) at birth helps to distinguish congenital infection from symptomatic postnatal infection, to which these NB are particularly vulnerable [28].

The 2026 AAP list, which defines the expanded testing recommended in the USA, includes term and preterm NB with maternal HIV or CMV infection in pregnancy, failed hearing screening, compatible abnormal CNS imaging, prenatal or postnatal imaging with intra-abdominal calcifications, hepatomegaly, splenomegaly, petechial rash, raised aminotransferases or conjugated bilirubin, thrombocytopenia without another explanation, birth weight below the 10th centile for gestational age or head circumference below the 10th centile [3]. In preterm infants who do not undergo hearing screening by 21 days, the AAP suggests considering testing before 21 days, which also makes it possible to diagnose a postnatal infection later [3].

4.3. In the infant and child beyond the neonatal period

Confirmed SNHL, microcephaly or intracranial calcifications discovered after 3 weeks of life raise the problem of retrospective diagnosis: PCR on urine or saliva no longer distinguishes congenital from postnatal infection, and the informative test becomes CMV PCR on the dried blood spot collected for newborn screening [3, 12, 19]. Testing for cCMV is part of the routine investigation of cases of deafness confirmed after newborn hearing screening [21]. In a multinational surveillance study (United Kingdom and Ireland, Canada, Australia; 2016–2018) of 118 infants with severe microcephaly (head circumference ≤−3 standard deviations), the cause was genetic in 50%, acquired in 18% (infection, ischaemia or hypoxia, alcohol, placental insufficiency) and unknown in 32%, and CMV testing was performed in only 48% [49].

4.4. Red flags and emergency signs

Severe thrombocytopenia with petechiae or bleeding, marked hepatic dysfunction or liver failure, seizures, life-threatening multiorgan disease and gastrointestinal disease with enteral intolerance (for example, necrotising enterocolitis) require immediate assessment in a neonatal intensive care unit (NICU) and discussion with paediatric infectious diseases, a situation in which treatment may have to start intravenously [3, 28]. During antiviral treatment, fever in an infant with neutropenia is an emergency, and parents should know whom to contact [8].

4.5. Signs suggesting a genetic phenocopy

A clinical picture of cCMV without virological confirmation should raise the possibility of a phenocopy, especially when there is consanguinity, a previously affected sibling, cortical band-like calcifications with polymicrogyria, recurrent intracranial haemorrhages with congenital cataracts, sterile fevers, chilblains, irritability with loss of skills after a symptom-free interval, thrombocytopenia evolving into persistent pancytopenia or normal hearing in a severe encephalopathic picture [4, 12, 19]. Hearing is almost always normal in AGS, whereas SNHL is the most frequent sequela of cCMV [3, 4]. In the words of the GeneReviews chapter on AGS, congenital infections are the most common and the most important conditions to exclude in this differential, because misdiagnosis leads to incorrect counselling about the recurrence risk [4].

4.6. Newborn screening: universal, hearing-targeted or expanded

Between 80 and 90% of infected NB have no signs at birth and there is no surrogate marker to identify them, so early detection of late-onset SNHL depends on a screening strategy [3]. Three strategies are distinguished: hearing-targeted screening, which tests all NB who fail hearing screening; expanded testing, which adds risk factors and clinical or laboratory signs; and universal screening, on saliva or dried blood spot [3].

In Minnesota, in a tertiary paediatric hospital (132 patients, 2021–2024), age at the first audiology visit after diagnosis fell from 0.71 ± 1.8 years before 2022 to 0.07 ± 0.06 years with universal screening (p=0.02), with more diagnoses of mild SNHL and no difference in severe to profound SNHL [53]. In south-eastern Minnesota, the observed birth prevalence was about 3 times higher with hearing-targeted screening and about 18 times higher with universal screening than before, which shows that most NB with cCMV are not identified without systematic screening [54]. In the USA, the AAP estimates that 0.03–0.09% of screened NB would benefit from early detection of late-onset SNHL, or 0.03% (about 1200 children per year, 1 in 3500) if only SNHL before 24 months with reasonable or good evidence is counted [3]. The estimated cost of universal saliva screening is about 50 dollars per NB, with 2000–10,000 dollars (2016 values) per infected NB detected and 975–90,038 dollars per deaf or hard-of-hearing child identified; in the United Kingdom, hearing-targeted screening cost 6683 pounds (2011) per infant with SNHL due to cCMV [3].

Positions diverge. In 2026 the AAP recommends a hearing-targeted programme combined with expanded testing and considers the data insufficient to recommend universal screening; the 2022 proposal to include cCMV in the US Recommended Uniform Screening Panel did not meet the evidence standards [3]. The American Cochlear Implant Alliance, the National CMV Foundation and the American Academy of Otolaryngology–Head and Neck Surgery (AAO-HNS) advocate universal screening as necessary, a position shared by a 2026 commentary in Pediatrics [3, 10, 55]. The 2017 international group recommended considering universal newborn screening [24]. The UK National Screening Committee (UK NSC), after an evidence review in 2025 (217 studies identified; saliva and dried blood spot probably with detection of more than 90% of cases; no study compared screening with current practice) and a public consultation between 14 November 2025 and 6 February 2026, proposed on 25 June 2026 to maintain the recommendation not to screen NB, because of insufficient evidence of benefit over existing clinical management [56, 57]. The final decision of that meeting was not located in the sources consulted.

5. Diagnosis: Criteria, Tests and Algorithms

5.1. Diagnosis of maternal infection

Serological status is established with latest-generation IgG assays, with a sensitivity of 97–100% and a specificity of 96–100%; discordant results, which occur in 1.0–2.6% of samples and mainly with IgG below twice the cut-off, are repeated with a second assay or in a reference laboratory and, if they persist, are considered equivocal and treated as negative [8]. Latest-generation IgM has a sensitivity above 98% for primary infection occurring in the previous month, which falls to 86–97% and 51–90% in the 2nd and 3rd months, but specificity is poor: IgM persists in 50–80% of women for up to 6 months after primary infection, may appear in non-primary infections and through cross-reactivity, and in routine 1st-trimester screening it is positive in 0.9–5.7% of pregnant women, with a positive predictive value of 16.4% for recent primary infection [8]. The SMFM reports that fewer than 10% of women with positive IgM transmit the infection, versus 30–50% of those with documented seroconversion; in a cohort of 2477 women referred for positive IgM, 55% were not confirmed by immunoblot, and of the 514 with confirmed IgM and low or moderate avidity 25% had an infected child [43].

IgG avidity is the test that dates the infection. It excludes recent (<3 months) primary infection with a sensitivity of 94–100% and semi-recent (3–6 months) primary infection with 80–90%; in recent primary infection avidity is low in 85–90% and intermediate in 10–15%, and at 6 months intermediate avidity persists in 20–40% [8]. Pregnant women in the 1st trimester with positive IgM and confirmed low or intermediate avidity are eligible for secondary prevention [8]. Thresholds depend on the assay: in a French cohort of 561 women (2012–2024), without treatment, transmission was 34% (VIDAS) and 31% (LIAISON) with low avidity and 14% and 22% with intermediate avidity, and a VIDAS index ≥0.60, or ≥0.50 before 12 weeks, was associated with 0% and 5% transmission; the authors propose considering avidity above 0.250 as high in the LIAISON assay [58]. PCR on whole blood has a sensitivity of 100% in the first 2 weeks and 88–97% at 4 weeks after primary infection, but DNAaemia persists for up to 1 year and cannot be used for dating; PCR on urine is positive in 90%, 95%, 70%, 45% and 12% of primary infections that occurred 1, 2, 6, 12 and 24 months earlier [8]. There is no valid test to identify, among seropositive women, those who will transmit: IgM is detected in 0–25%, a rise in IgG in 0–22% and DNAaemia in 24–66% [8].

The European consensus algorithm for 1st-trimester serology interprets the four combinations of IgG and IgM requested as early as possible [8].

The same document warns of two pitfalls: in acute infection with low IgG and high IgM, avidity may be falsely high with some assays, and with IgG below twice the cut-off avidity may be falsely low or intermediate; in both cases, a second sample clarifies the kinetics [8]. When the timing is unknown, retrospective serology on stored pregnancy sera helps to date it (grade A) [8].

5.2. Antenatal serological screening

Most European countries do not perform routine serology in pregnancy, although a French cost-effectiveness study concluded that universal screening combined with valaciclovir would be cost-effective [8]. The SMFM does not recommend routine screening of all pregnant women (grade 1B), and the American College of Obstetricians and Gynecologists (ACOG) and the Society of Obstetricians and Gynaecologists of Canada (SOGC) also do not recommend universal screening [3, 43]. A 2009 US analysis concluded that screening would only be cost-effective if prevention reduced transmission by more than 47% [3]. In Italy, the Istituto Superiore di Sanità and the Ministry of Health recommended in December 2023 (published in 2026) universal serological screening at the first antenatal visit, with monthly repetition in seronegative women until 24 weeks [59]. A decision model for 400,000 pregnancies per year estimated that screening up to 13+6 weeks would identify 910 primary infections and prevent 92% of symptomatic cCMV (183 versus 15) and 70% of TOP (33 versus 10), with a cost per symptomatic case prevented of about 45,500 euros [60]. In 2026 the UK NSC decided to commission a full formal evidence review on antenatal screening, which the evidence map considered promising because of the accuracy of IgG, IgM and avidity in the 1st trimester and the reduction in transmission with oral treatment, supported by a single published randomised trial [56]. The comparison of obstetric guidelines published in 2025 summarises the divergence: SMFM, the Royal College of Obstetricians and Gynaecologists (RCOG) and the Royal Australian and New Zealand College of Obstetricians and Gynaecologists (RANZCOG) against universal screening, SOGC in favour of early testing only in high-risk women where avidity is available, and ECCI in favour of routine screening up to 16 weeks [61].

5.3. Diagnosis of fetal infection

CMV PCR on AF is the reference standard for fetal diagnosis; cordocentesis has similar sensitivity and specificity but more complications [8, 43]. The timing of amniocentesis is the main point of divergence between guidelines. The 2024 European consensus, based on a study of 2706 cases, considers PCR reliable from 17+0 weeks provided amniocentesis is performed at least 8 weeks after primary infection (grade B), with specificity close to 100% and sensitivity of about 87–95% [8]. The SMFM recommends amniocentesis after 21 weeks and more than 6 weeks after maternal infection (grade 1C), with specificity of 97–100% and sensitivities of 45–80%, and proposes deferring until 21 weeks or repeating a negative early amniocentesis [43]. The classic studies collected in the supplements to the European consensus show why gestational age matters: sensitivity of 45% below 21 weeks (Donner, 1994; 11 cases), 30% below and 74% above 21 weeks (Liesnard, 2000; 237 cases) and 66.7% below and 96% above 21 weeks (Enders, 2001; 242 cases) [8]. In a meta-analysis, 8% of NB infected after a negative amniocentesis performed at the appropriate time had no sequelae at 2–3 years, which is attributed to late transmission [8]. Amniocentesis does not determine severity, and a positive or negative PCR on AF should be confirmed in the NB, because false positives and false negatives have been described [28, 43].

Ultrasound does not diagnose fetal infection, but serial ultrasound and 3rd-trimester fetal MRI provide prognostic information in infected fetuses (grade A) [8, 43]. In an infected fetus with normal ultrasound and MRI, the negative predictive value for moderate to severe sequelae approaches 100%, with a residual risk of unilateral SNHL of 17%; isolated extracerebral findings are associated with a 30% risk of sequelae, and severe brain anomalies with a poor prognosis [8]. In 2016 the SMFM considered the data insufficient to recommend routine fetal MRI, a position predating the evidence gathered by the European consensus [8, 43]. The Cannie fetal MRI scoring system, adopted by the European consensus, grades the risk [8].

No fetal biomarker has been validated; AF proteomics and cellular immunity remain under investigation [3]. In a single-centre retrospective study, the predictors of neonatal disease after primary infection were young maternal age, a high-risk occupation, documented IgG seroconversion and fetal anomalies on MRI, but not fever, abnormal fetal ultrasound or low avidity [3]. AF viral load has limited evidence as a prognostic factor [56].

5.4. Neonatal diagnosis

The diagnosis of cCMV requires detection of the virus in a sample collected in the first 3 weeks of life (21 days), because after that period detection does not distinguish congenital infection from intrapartum or postnatal infection [3, 8]. Real-time PCR has replaced culture, and PCR on urine collected by 21 days is the reference standard, with a sensitivity of 100% and a specificity of 99%; a single negative urine excludes infection, and the 2017 consensus recommends collecting it ideally in the first 14 days [3, 28]. Saliva is an alternative in term and late preterm NB, collected immediately before a feed in breastfed infants, and a positive saliva is confirmed on urine by 21 days [3, 28]. In the US multicentre CHIMES study, PCR on liquid saliva (17,662 NB) had a sensitivity of 100% (95% CI 95.8–100), specificity of 99.9%, positive predictive value (PPV) of 91.4% and negative predictive value (NPV) of 100%, and dried saliva (17,327 NB) a sensitivity of 97.4% (90.8–99.7) [52]. Repeat testing at 5 weeks was negative in 7.5% of initial positives, probably because of contamination with breast milk, and in very low birth weight preterm infants saliva has a sensitivity of 89.4% (95% CI 76.9–96.5) [3]. Overall, saliva compared with urine has a sensitivity of 93–100%, NPV of 98–99%, specificity of 91–99.7% and PPV of 49–73% [8].

PCR on the dried blood spot (DBS) is the retrospective reference standard, but the viral load in neonatal blood is lower than in saliva and urine and may be undetectable [8]. Sensitivity has changed with the methods: 28.3–34.4% in the 2007–2008 CHIMES study, which concluded that the assays of the time were inadequate for screening [62]; 73.2–76.8% for each laboratory and 85.7% for the combined result, versus 92.9% for saliva, in Minnesota (2016–2019; 12,554 NB; 56 cases confirmed on urine) [63]; and about 84% in a meta-analysis, with wide variation between laboratories [28]. These values measure the proportion of urine-confirmed cases that the DBS detects, not the ability to predict outcomes; a negative DBS does not exclude cCMV [28, 63]. IgM should not be used in the NB; a negative IgG in the mother or the NB at birth excludes cCMV, but a positive IgG decides nothing [8].

5.5. Assessment after confirmation of infection

The European consensus recommends the same neonatal assessment whatever the type of maternal infection: anthropometry (weight, length, head circumference) and complete physical examination, full blood count, aminotransferases and total and conjugated bilirubin, ophthalmological assessment and audiological assessment at birth (grade A) [8]. Hearing assessment is performed by ABR, because otoacoustic emissions do not detect hearing loss of retrocochlear origin [28]. Cranial ultrasound is the first-line investigation; MRI, which does not require sedation in the NB, detects white matter abnormalities and cortical malformations missed by ultrasound, and is indicated in all NB with clinical manifestations, SNHL, chorioretinitis or abnormal ultrasound, and may be performed when primary infection occurred in the 1st trimester or is of unknown timing (grade B) [8, 28]. The AAP recommends MRI without contrast if the neurological examination or ultrasound is abnormal, with some experts requesting it in all disease, diagnostic audiological assessment by 12 weeks, ophthalmology and paediatric infectious diseases for all disease and for isolated SNHL, and does not consider routine ophthalmology necessary in infection without signs because of the low yield [3]. Renal function should be assessed before treatment, and baseline blood viral load measured, although its absence does not exclude infection [28]. Lumbar puncture is not routinely recommended (grade B), because CMV DNA is detected in cerebrospinal fluid (CSF) in only about 13–15% of NB with CNS involvement [8].

The European consensus uses a combined postnatal cranial ultrasound and MRI score (Alarcon) to guide treatment: scores of 2 or 3 justify treatment and a score of 1 justifies expert opinion [8].

5.6. Classification of severity

The 2026 AAP proposes four categories that determine the treatment decision (section 7); the classification derives from the 2017 European consensus, which defined severe, moderate and mild disease [3, 28].

5.7. Limitations of the tests

5.8. Genetic investigation when cCMV is not confirmed or does not explain the picture

When PCR on urine or saliva by 21 days, or on the dried blood spot after that period, is negative in an infant with a suggestive picture, or when the extent of the lesions is not compatible with the timing of maternal infection, investigation turns to the phenocopies [12, 19]. Neuroimaging is central: CT has the highest sensitivity for calcification and MRI may miss even extensive calcifications, so CT is considered in a child with undiagnosed neurological disease, especially white matter disease; gradient echo and susceptibility-weighted sequences are more sensitive for haemorrhage and mineralisation, and repeat imaging distinguishes resolving haemorrhage from calcification [19]. The initial investigation includes urea and electrolytes, calcium, phosphate and parathyroid hormone, congenital infection serology, CMV PCR on the Guthrie card, urinary organic acids and amino acids, plasma lactate and amino acids and CSF with cells, protein, lactate, interferon α and pterins, followed by genomic analysis (microarray, panel or exome or genome sequencing) unless a specific monogenic disorder is suspected, with DNA storage [19]. When AGS is suspected, the blood interferon signature (qPCR of interferon-stimulated genes; positive in 90% of patients with a proven variant), CSF interferon α activity (normal <2 IU/mL) and CSF lymphocytosis (>5 cells/mm³) support the diagnosis; the signature is raised throughout the interferonopathy group and does not identify the gene [4, 12]. C26:0-lysophosphatidylcholine, used in newborn screening for X-linked adrenoleukodystrophy, is raised at birth in patients with AGS [33].

The choice of molecular method must take specific limitations into account. OCLN has a pseudogene 1.5 Mb away and a spectrum rich in copy number variants, so a standard panel or exome may miss the diagnosis; sequencing is combined with copy number analysis [12, 64]. In SAMHD1, deletions and duplications account for up to 30% of alleles [4]. In deafness, molecular testing is the standard of care, with a multigene panel or exome or genome sequencing; single-gene testing is rarely useful, microarray alone has a low yield, and a genetic diagnosis is obtained in up to 60% of children with severe to profound bilateral congenital SNHL [5, 65]. The GJB6 deletion is not detected by sequencing of the GJB2 coding region, and mitochondrial DNA is tested when there is maternal inheritance or aminoglycoside exposure [21]. In branchio-oto-renal syndrome, complex EYA1 rearrangements escape sequencing [21]. The report should identify the gene, the transcript with version, the variant in HGVS, zygosity and the class according to the criteria of the American College of Medical Genetics and Genomics and the Association for Molecular Pathology (ACMG/AMP); a variant of uncertain significance requires segregation studies in the parents, in recessive disorders to confirm the phase of the two variants, and does not by itself support recurrence counselling. A negative genetic result does not exclude a phenocopy, and a positive result does not exclude coexisting cCMV [3].

6. Differential Diagnosis

6.1. Comparative table

6.2. Other congenital infections

Intracranial calcifications are common but not constant in congenital infection and have been described after CMV, rubella, Zika, toxoplasma, parvovirus, varicella, herpes, lymphocytic choriomeningitis virus and HIV [19]. Diagnosis is simple in active neonatal infection with identification of the agent, and more difficult after the neonatal period, when the radiological phenotype becomes the main guide [19]. In congenital Zika syndrome, calcifications are predominantly pericortical, unlike the periventricular calcifications of cCMV, and the external phenotype of the fetal brain disruption sequence is rarely seen in severe CMV infection [66, 67]. Zika virus spread through the Americas after its first identification in Brazil in early 2015, and the causal relationship with microcephaly and severe brain anomalies was established in 2016 [23, 71]. In Brazil, the prevalence of congenital Zika syndrome was 3.8 per 10,000 live births in 2015 and 3.1 per 10,000 in 2016 [67]. In congenital toxoplasmosis, transmission after maternal seroconversion rises from 15% (95% CI 13–17) at 13 weeks to 44% at 26 weeks and 71% at 36 weeks, with about 1 in 5 infected infants showing signs [68], the same inverse pattern between transmission and severity described for cCMV. Maternal rubella between 2 and 10 weeks damages the fetus in up to 90% of cases, and SNHL may be detected later in children apparently normal at birth [69]. The comparative table in the Brazilian book on genetic syndromes summarises the differences between Zika and the other infections: CMV with microcephaly, hearing impairment, strabismus and hypertonia, but periventricular calcifications; toxoplasmosis with hepatosplenomegaly, jaundice, anaemia, hydrocephalus, eosinophilia and microphthalmia; rubella with myocarditis, glaucoma, pigmentary retinopathy, microphthalmia and cataract; herpes simplex with nystagmus, Horner syndrome, microphthalmia and chorioretinitis; and HIV with acquired microcephaly, hepatosplenomegaly and opportunistic infections [67]. Neonatal cholestasis with paucity of bile ducts has among its causes cCMV and other congenital infections, which includes it in the differential diagnosis of Alagille syndrome [72].

6.3. Type I interferonopathies: Aicardi-Goutières syndrome and pseudo-TORCH types 2 and 3

AGS is the most important Mendelian mimic of congenital infection [19]. About 20% of patients, mainly with biallelic TREX1 variants, present at birth with neurological abnormalities, hepatosplenomegaly, raised aminotransferases and thrombocytopenia, a picture reminiscent of congenital infection and repeatedly diagnosed as such; the remaining 80% have a normal pregnancy, delivery and neonatal period and, after the first weeks, develop a subacute encephalopathy with extreme irritability, intermittent sterile fevers, loss of skills and deceleration of head growth [4, 12]. GeneReviews includes as an exclusion criterion evidence of prenatal or perinatal infection by CMV, toxoplasmosis, rubella, herpes simplex, Zika or HIV [4]. Three features help to distinguish the two situations: hearing is almost always normal in AGS, polymicrogyria has never been described in AGS and AGS calcification predominates in the basal ganglia and has a paraventricular, not truly periventricular, distribution [4]. The prenatal form with thrombocytopenia, anaemia and hepatic dysfunction may resolve within weeks and carries profound delay with a higher risk of death in the first year [33]. The microcephaly–intracranial calcification syndrome described by Reardon in 1994 corresponds, in most cases, to AGS [4]. The 9 AGS genes, treatment with JAK inhibitors and recently proposed variants, such as PTPN1 (2026 preprint, 9/53 undiagnosed AGS patients with median onset at 1.75 years) [37], belong to the domain of Aicardi-Goutières syndrome and the type I interferonopathies.

USP18 deficiency (pseudo-TORCH type 2), described in 2016 in 5 patients from 2 families, causes microcephaly, ventriculomegaly, brain calcification and, sometimes, systemic signs at birth similar to the sequelae of congenital infection without an infectious agent; autopsy showed innate inflammation with calcification and polymicrogyria [70]. In 2020, a NB with hydrocephalus, necrotising cellulitis, systemic inflammation and respiratory failure due to a homozygous USP18 splice variant made a rapid and sustained recovery with ruxolitinib, a single, off-label case [34]. STAT2 gain of function (pseudo-TORCH type 3) was described in two siblings with severe early-onset autoinflammatory disease [35]. The practical message of the GeneGuidance curation is that an encephalopathic NB with intracranial calcifications and haemorrhage and a negative TORCH screen should be investigated genetically (OCLN, STAT2, JAM3, USP18), and not treated indefinitely for a presumed occult infection [12].

6.4. Phenocopies due to injury of tight junctions and small vessels

Band-like calcification with simplified gyration and polymicrogyria (pseudo-TORCH type 1) was associated in 2010 with intragenic deletions and variants of OCLN in 9 patients from 6 families [38]. The presentation is severe, at birth or shortly afterwards, with seizures, feeding difficulties, microcephaly, tetraparesis and bulbar palsy, and the imaging is characteristic: generalised brain malformation with primitive sulcation, a continuous or semi-continuous symmetrical band of cortical calcification on CT and symmetrical thalamic and, sometimes, pontine calcification [19]. The pattern differs from cCMV, which is the most common cause of the association between intracranial calcification and polymicrogyria [19]. The deletion of exons 3–4 of OCLN is a founder variant in the Egyptian population [12]. The homozygous JAM3 splice variant described in a consanguineous family from the United Arab Emirates causes severe haemorrhagic destruction of the brain, subependymal calcification and congenital cataracts; massive intracranial haemorrhage distinguishes it from OCLN pseudo-TORCH [39]. The cataracts may be seen at 23 weeks, but a normal fetal brain ultrasound does not exclude the disease, and auditory neuropathy has been added to the phenotype [12]. In COL4A1/COL4A2 disease, porencephaly and periventricular leukomalacia may mimic an acquired destructive lesion [19, 73].

6.5. Leukoencephalopathies with cysts and calcifications

RNASET2 cystic leukoencephalopathy without megalencephaly combines severe psychomotor delay, microcephaly, seizures and SNHL with bilateral anterior temporal subcortical cysts, multifocal white matter lesions, ventriculomegaly and calcifications, and is essentially indistinguishable from the brain injury of cCMV without signs at birth [12, 19]. Because the disease also causes deafness, the discriminating test in an infant beyond the neonatal period is CMV PCR on the newborn screening dried blood spot, complemented by audiology, ophthalmological examination (chorioretinitis points to infection) and, if AGS is in the differential, CSF and the interferon signature [12]. RMND1 leukoencephalopathy, with temporal cysts and deafness, is another mimic [12]. Coats plus (CTC1) and leukoencephalopathy with calcifications and cysts (SNORD118) share dense calcification of the basal ganglia, thalami, cerebellum and deep cortex with expansile cysts, the former being distinguished by retinal, bone and gastrointestinal abnormalities [19].

6.6. Microcephaly and "fetal brain arrest"

Several rare autosomal recessive disorders can produce a phenotype compatible with the fetal brain disruption sequence of Zika: AGS, pseudo-TORCH and variants of JAM3, NDE1 and ANKLE2; the term "fetal brain arrest" has been proposed for some of these cases [66]. In the series of 12 individuals with ANKLE2 variants, all had microcephaly with z ≤−3 (congenital in 9), simplified gyration and agenesis of the corpus callosum, and 6 had hyper- or hypopigmented macules [12]. After exclusion of syndromic and environmental causes, with a structurally normal MRI, the empirical recurrence risk of primary microcephaly is 15–20%, and the Oxford Desk Reference explicitly warns against genetic disorders that mimic environmental causes, such as pseudo-TORCH [74].

6.7. Genetic causes of deafness

In developed countries, about 65% of prelingual deafness is genetic and 35% acquired, although the same GeneReviews chapter gives 80% genetic causes in another section [5]. Severe to profound deafness affects about 1 in 1000 infants at birth or in the prelingual period, and a further 2–3 in 1000 children have moderate or progressive deafness; of the genetic causes, about 30–35% are syndromic and 65–70% non-syndromic, and of these 80% are autosomal recessive [5, 21]. GJB2 (connexin 26, DFNB1) accounts for up to 50% of non-syndromic recessive prelingual deafness, and the 342 kb deletion involving GJB6 is the second cause in Mediterranean populations [21]. Mitochondrial variants are present in up to 30% of families with deafness in at least 2 generations through the maternal line, but in only 1% of sporadic early non-syndromic deafness [21]. In SLC26A4 deafness, an enlarged vestibular aqueduct is not specific, because cCMV also causes malformations of the temporal bone; SLC26A4 testing is not indicated when the anomalies do not follow the typical pattern [22]. Auditory neuropathy, with otoacoustic emissions present and absent or abnormal ABR, suggests OTOF in the term NB, and the intracochlear gene therapy approved for this form cannot be performed in an already implanted ear [5, 21, 42]. Deafness associated with severe jaundice or perinatal factors is, as a rule, accompanied by neurological and learning problems [21].

Three practical situations summarise the relationship between cCMV and genetic deafness. In a NB with SNHL and a positive CMV PCR by 21 days, cCMV is the most likely cause, but genetic assessment remains indicated because the two may coexist [3]. In a child with SNHL diagnosed after 3 weeks, PCR on the neonatal dried blood spot is the only test that can attribute the deafness to cCMV, and a negative result does not exclude it [28, 63]. In a family with deafness of unexplained cause, the empirical recurrence risks of genetic deafness (section 8) do not apply if cCMV is confirmed, which gives CMV testing direct value in counselling [5].

7. Treatments and Risk Management Strategies

7.1. General principles

No antiviral is authorised for the treatment of cCMV, either in the USA or in Europe; valganciclovir and ganciclovir are used off-label in the NB, and high-dose valaciclovir is used off-label in the pregnant woman [28, 75–77]. The treatment decision should be shared with the parents, with information on the efficacy and toxicity data from the trials, and discussed with paediatric infectious diseases; the European consensus recommends offering inclusion in the international CCMVNET registry and its monthly virtual clinic [8]. Treatment is therefore of three types: neonatal antiviral therapy to reduce the progression of SNHL and improve neurodevelopment, secondary prevention in the pregnant woman with early primary infection, and non-pharmacological intervention (audiological, rehabilitative and educational) for all children with sequelae.

7.2. Neonatal antiviral treatment: indications

Ganciclovir is a nucleoside analogue of guanosine that inhibits CMV DNA synthesis; valganciclovir is its valine ester prodrug, converted into ganciclovir by intestinal and hepatic esterases [3]. Oral valganciclovir is the treatment of choice, and intravenous ganciclovir is reserved for when the enteral route cannot be used or for very severe cases, switching to the oral route as soon as possible [8, 28]. The table compares the indications in the three reference guidelines.

The European consensus operationalises duration by the presence of CNS involvement, which includes SNHL, chorioretinitis, clinical symptoms and abnormal brain imaging (6 months), versus the absence of CNS involvement, for example isolated hepatitis or thrombocytopenia (6 weeks or longer, depending on the response) [8].

7.3. Dosage, formulations and eligibility

The dose of 16 mg/kg twice daily derives from a pharmacokinetic and pharmacodynamic study in 24 NB with cCMV and CNS involvement: between 7 days and 3 months, the oral solution at that dose gave an exposure (median area under the curve from 0 to 12 h of 23.6 mcg·h/mL; n=6) comparable to that of intravenous ganciclovir 6 mg/kg twice daily (25.3 mcg·h/mL; n=18) and to that of adults on 900 mg twice daily [75]. The US label states that safety and efficacy have not been established in cCMV and labels paediatric dosing only for prevention in kidney and heart transplantation, with a single daily dose calculated as 7 × body surface area × creatinine clearance (modified Schwartz), maximum 900 mg [75]. The UK summary of product characteristics (SmPC) describes the 6-week versus 6-month trial in 109 infants, but states that this indication is not currently recommended for valganciclovir, because the design and results of the studies are too limited for conclusions on efficacy, safety and dosage [76].

7.4. Monitoring and adverse effects

Neutropenia is the main adverse effect and is concentrated in the first 4–6 weeks [3]. In the 2003 trial, grade 3–4 neutropenia occurred in 29/46 (63%) NB treated with intravenous ganciclovir, with the dose adjusted in 14 and stopped in 4, versus 9/43 (21%) controls [3, 26]. In the 2015 trial it occurred in 19% in the first 6 weeks of valganciclovir and, in the following 4.5 months, in 21% with valganciclovir versus 27% with placebo (p=0.64), which shows that prolongation does not add haematological toxicity [9]. In the CCMVNET registry, severe neutropenia (<0.5 × 10⁹/L) affected 18.4% of treated infants, associated with intravenous ganciclovir for more than 14 days, prematurity (<36 weeks) and lower neutrophil counts before treatment [8]. Hepatotoxicity in up to 30% of those treated with ganciclovir and thrombocytopenia in a similar proportion have been described; with valganciclovir, mild aminotransferase elevations were not clinically or statistically significant compared with placebo, and the abnormalities resolved after discontinuation [3, 28]. Nephrotoxicity is possible with other nephrotoxic drugs [3]. Infertility, mutagenicity and carcinogenicity have been demonstrated in animals and not in humans; the US label has a boxed warning for haematological toxicity, impairment of fertility, fetal toxicity, mutagenesis and carcinogenesis [3, 75]. There are no data on long-term effects in treated children, and parents should be informed of this uncertainty [8, 28].

In a 2023 survey, 56% of US paediatric infectious disease specialists followed the monitoring proposed by the AAP [3]. Parents should know what to do in the event of fever or illness during neutropenia [8]. Contraception and breastfeeding appear on the label for adult patients: breastfeeding not recommended during treatment, effective contraception in women during and for ≥30 days after, and condoms in men during and for ≥90 days after [75].

7.5. Non-pharmacological interventions

Hearing aids serve mild to severe deafness, with moulds replaced as the canal grows, and bone conduction devices some selected cases [5, 21]. Cochlear implantation is considered in children with severe to profound deafness from 9 months, and implantation before 2 years may allow oral language indistinguishable from that of peers at school age [5]; in candidates with severe to profound SNHL due to cCMV, early implantation is associated with better acquisition of oral language [3]. Implant outcomes in GJB2 deafness are significantly better than in deafness of environmental cause [5]. The AAP recommends offering early intervention to all children with cCMV disease or isolated SNHL [3]. Physiotherapy, speech therapy and occupational therapy are part of the care of children with motor, language or communication sequelae, and children with deafness may also communicate through sign language [30, 32].

7.6. Secondary prevention in pregnancy: valaciclovir

Valaciclovir is the prodrug of aciclovir and has no approved indication for CMV; the US label (Valtrex®) authorises it for herpes labialis, genital herpes, herpes zoster and paediatric varicella, with a maximum dose of 1 g three times daily in adults, so the 8 g/day used in pregnancy is about 2.7 times that dose and constitutes off-label use [77]. The double-blind randomised trial at Rabin Medical Center (Israel; NCT02351102; November 2015 to October 2018) included pregnant women with periconceptional or 1st-trimester primary infection, treated with valaciclovir 8 g/day in 2 doses (4 g twice daily) or placebo until amniocentesis at 21–22 weeks; of the 100 randomised, 90 were analysed per protocol [27]. Vertical transmission was 5/45 (11%) with valaciclovir and 14/47 (30%) with placebo (p=0.027; odds ratio, OR, 0.29; 95% CI 0.09–0.90), and in 1st-trimester infection 2/19 (11%) versus 11/23 (48%) (p=0.020), without clinically significant adverse events [27]. The 2024 individual participant data meta-analysis (3 studies, n=527) estimated an adjusted OR (aOR) of 0.34 (95% CI 0.18–0.61) for vertical transmission, 0.30 (0.19–0.47) for neonatal infection and 0.23 for TOP for severe fetal findings associated with CMV, with serious adverse effects in 2.1%; the later treatment was started, the worse the outcomes [79]. The 2025 update (4 studies, n=860) obtained an aOR of 0.39 (0.25–0.59) for transmission, 0.30 (0.13–0.68) for periconceptional infection and 0.47 (0.28–0.78) in the 1st trimester [80]. A meta-analysis cited by the AAP recorded pooled adverse events in 3.17% (95% CI 1.24–5.93) and reversible acute kidney injury in 1.71% (0.41–3.39) [3].

According to the European consensus, the 8 g/day dose is given as 2 g four times daily (grade D), because in one series acute kidney injury occurred in 2/50 (4%) pregnant women on 4 g twice daily and in 0/173 on 2 g four times daily; in the meta-analysis, mild to moderate reversible renal impairment occurred in 3 cases (2%), 1 on 2 g four times daily and 2 on 4 g twice daily, and mild effects (nausea, headache) in 21% [8]. A 2026 report described a new acute kidney injury in a healthy pregnant woman with 1st-trimester primary infection and estimated the risk at about 1% [81]. The US label mentions thrombotic thrombocytopenic purpura and haemolytic uraemic syndrome, sometimes fatal, in patients with advanced HIV disease and transplant recipients treated with 8 g/day in clinical trials [77]. Decades of use of valaciclovir and aciclovir in pregnancy for herpes simplex and varicella have not identified a risk of major malformations [8, 77].

The most recent data reinforce and qualify this effect. In the French cohort of 561 women (2012–2024), valaciclovir reduced transmission to 2%, 5% and 16% in preconceptional, periconceptional and 1st-trimester primary infections (p=0.034, 0.003 and <0.001) [58]. In Tübingen (October 2023 to June 2025), among 85 treated pregnant women with primary infection before 14 weeks (median gestational age at diagnosis 7.4 weeks), transmission was 7/85 (8.2%), all in women without anti-gB2 IgG antibodies (7/55; 12.7%) versus 0/30 with these antibodies (p=0.048), a potential risk marker not yet validated [83].

7.7. Treatment of the infected fetus

In a French open-label, single-arm phase II trial, 43 pregnant women with an infected fetus and extracerebral or mild cerebral ultrasound signs received valaciclovir 8 g/day from diagnosis (median 25.9 weeks) until delivery or TOP (median duration 89 days); 34 NB were born asymptomatic and remained so at 12 months, fetal blood viral load decreased and platelets increased, and compared with a historical cohort the proportion of asymptomatic NB was 82% versus 43% without treatment; the earlier placebo-controlled trial had failed to recruit [82]. In a case–control study, valaciclovir 2 g four times daily from amniocentesis until delivery reduced symptomatic NB from 66% to 18% [8]. In a retrospective Italian cohort of 77 infected fetuses without major malformations, fetal MRI remained normal in 73.3% of the 15 treated versus 37.1% of the 62 untreated (OR 4.67; 95% CI 1.29–16.9) [47]. These are non-randomised data with historical or concurrent controls, and the European consensus gives this indication grade C [8]. Prenatal valganciclovir is not recommended [8]: ganciclovir caused maternal and fetal toxicity, embryo-fetal mortality and teratogenicity in animals at exposures twice the human exposure, with no human data [75].

7.8. CMV hyperimmune globulin

An uncontrolled 2005 study reported a reduction in transmission from 40% to 16%, but subsequent randomised trials were negative [84]. In the Italian CHIP trial (phase 2, double-blind; 123 evaluated pregnant women with primary infection at 5–26 weeks), hyperimmune globulin (HIG) every 4 weeks until 36 weeks resulted in congenital infection in 30% (18/61) versus 44% (27/62) with placebo (difference of 14 percentage points; 95% CI −3 to 31; p=0.13), with adverse obstetric events in 13% versus 2% [84]. In the US trial of the NICHD Maternal-Fetal Medicine Units network (2012–2018; 206,082 pregnant women screened, 712 primary infections, 399 randomised), stopped for futility, the primary outcome occurred in 22.7% with monthly HIG and 19.4% with placebo (relative risk, RR, 1.17; 95% CI 0.80–1.72), with an unfavourable trend for death (4.9% versus 2.6%), preterm birth (12.2% versus 8.3%) and weight below the 5th centile (10.3% versus 5.4%) [85]. The European consensus recommends against HIG 100 IU/kg every 4 weeks (grade A) and accepts considering 200 IU/kg every 2 weeks in very recent 1st-trimester primary infection (grade C), based on a case–control study with a 70% reduction in transmission [8]. The article reporting the US trial describes the dose as 100 mg per kilogram, whereas the European consensus expresses it in international units per kilogram [8, 85].

7.9. Primary prevention

Hygiene considerably reduces the risk of primary infection in several studies, and the European consensus recommends advising it to all women, whatever their serological status, from before conception and during the 1st trimester, especially to known seronegative women (grade B), with no proof of benefit in seropositive women and little evidence on the sexual route [8]. The AAP measures are washing hands after contact with objects contaminated by saliva or urine (nappies, dummies) and after feeding a young child or wiping their nose or face, not sharing cups, cutlery or toothbrushes with young children, not putting dummies in one's mouth and avoiding kissing young children on the mouth [3]. Educational interventions (films, leaflets, calendar reminders) improve hygiene and are acceptable; in Europe, only 20–40% of pregnant women have heard of CMV and 10–15% of these know the preventive measures [8]. The AAP recommends that paediatricians offer people of reproductive age education on the risks, prevention and the support available [3].

7.10. Comparative regulatory status

The name, national authorisation and reimbursement of these medicinal products by INFARMED (Portugal) and ANVISA (Brazil) are not given in the sources consulted.

7.11. Main clinical trials

The neonatal trials are small, with follow-up of about 2 years, and otitis media with effusion confounds hearing assessment [3]. In the 2015 trial the benefit appeared only in secondary outcomes, and the comparison was between two treatment durations and not against no treatment [9]. No randomised trial included NB under 32 weeks or stratified by brain imaging, and the trials in infants without signs did not finish because of lack of recruitment or toxicity [3, 8]. Observational studies of isolated SNHL have no controls or use controls chosen by the parents: in Israel, 59 infants treated in the first 12 weeks improved in 69% of the 80 ears, and 21 infants with late-onset SNHL treated from 10.3 ± 7.8 months improved in 29/42 ears, both without controls [3]. In the longitudinal cohort of Lanzieri (76 children, 17 treated with intravenous ganciclovir for about 6 weeks, median follow-up 11–13 years), SNHL progressed with age in all groups, with no apparent lasting hearing benefit from 6 weeks; in a retrospective series of 16 treated children, 14 had clinically significant hearing decline after an average of 2.6 years [3]. For prevention in pregnancy, the only published randomised trial has 90 participants analysed per protocol and an intermediate outcome (PCR on AF) [27, 56]. Relative reductions should be read with the absolute rate: in 1st-trimester primary infection, with transmission of about 37% and sequelae in about a quarter of those infected, the absolute reduction in sequelae per treated pregnant woman is smaller than the relative reduction in transmission [8, 29]. There are no direct comparisons between trials, and the results should not be compared indirectly across different populations and outcomes.

7.12. Developments in the last 12 months and ongoing research

The search for trials with results of letermovir in pregnancy and for approved vaccines was negative in the databases consulted (ClinicalTrials.gov, Europe PMC, EMA, Drugs@FDA, 17-10-2026), which does not prove the non-existence of non-indexed data [86, 96].

7.13. Agents to avoid, pregnancy, lactation and particular populations

Valganciclovir and ganciclovir should not be used in pregnancy outside a trial, because of teratogenicity and embryo-fetal mortality in animals [8, 75]. Letermovir has no adequate human data in pregnancy and caused embryo-fetal toxicity with malformations in rats at exposures 11 times the human exposure [87]. Breastfeeding is not recommended for women taking valganciclovir [75]; breast milk transmits CMV, with silent infection in the term NB, and its effect in preterm infants is covered in section 9 [3]. In NB under 32 weeks or 1.8 kg, outside the trial populations, treatment requires expert opinion and, if possible, drug level monitoring [8]. Transition to adult care is not addressed in the sources consulted; SNHL may require lifelong audiological surveillance (section 11) [8].

7.14. Guideline recommendations and divergences

Critical appraisal of the guidelines with AGREE II (10 guidelines and 2 consensus statements up to April 2024) found agreement on the criteria for primary infection and on the reference standard for fetal diagnosis, but only two societies recommended routine valaciclovir, fetal surveillance was heterogeneous, and unconfirmed fetal infection and non-primary maternal infection were common gaps [99]. Japan published its first national guidelines in 2023, with 20 clinical questions, and Poland joint obstetric and infectious disease recommendations in 2025, whose full texts were not consulted [100, 101].

8. Reproductive Options and Family Counselling

8.1. cCMV is not inherited

cCMV has no Mendelian transmission, and counselling after an affected child rests on maternal immune status and prevention [1]. A woman who has had a child with cCMV is already seropositive: preconceptional immunity reduces the risk of cCMV in a subsequent pregnancy by about 69% (cCMV in 1.0% of the children of immune mothers, versus 3.0% of the children of initially seronegative mothers), but does not abolish it, because reactivation and reinfection can transmit the virus [3, 31]. Women who are seronegative in their first pregnancy have a 24-fold higher risk of having a subsequent child with cCMV, especially if the new pregnancy occurs within 2 years, because the older child of day-care age is the main source of primary infection [3, 8]. Once the fetus is infected, outcomes after non-primary and primary infection are similar, but the probability of transmission after non-primary infection is much lower (probably below 3.5%) [8]. There is no test that identifies seropositive women at risk of transmission, and the European consensus advises against testing women who were already seropositive with serology or PCR in pregnancy (grade B) [8].

8.2. Counselling before and during a new pregnancy

Hygiene counselling applies to all women, before conception and during the 1st trimester, with emphasis on seronegative women and those living with young children [3, 8]. In seronegative women, early serology repeated until 14–16 weeks makes it possible to identify a periconceptional or 1st-trimester primary infection in time to start valaciclovir before 16 weeks and to plan amniocentesis from 17 weeks, according to European practice (sections 5 and 7) [8]. Transmission after preconceptional primary infection is 5.5% in the meta-analysis, which underpins the information given to a woman with recent primary infection who wishes to become pregnant; the sources consulted do not establish a validated minimum interval between primary infection and conception [29]. With a diagnosis of fetal infection, the options include ultrasound and MRI surveillance with prognostic counselling based on the gestational age at infection and the findings (section 11), valaciclovir after discussion by an expert team and, depending on the legislation, TOP [3, 8]. A negative PCR on AF collected at the appropriate time allows the pregnant woman to be reassured, because late fetal infection does not leave long-term sequelae (grade A) [8].

8.3. Laboratory confirmation as a prerequisite for counselling

The point of contact between cCMV and medical genetics is the recurrence risk. Labelling a genetic phenocopy as congenital infection leads to counselling a low risk when it is 25% and deprives the family of prenatal diagnosis and preimplantation genetic testing [4, 12]. The Oxford Desk Reference recommends considering autosomal recessive pseudo-TORCH or microcephaly with intracranial calcifications whenever a baby with signs of congenital infection has no serological evidence and no virus identified [23, 68]. The infection should therefore be demonstrated virologically (PCR by 21 days or on the dried blood spot), and the clinical and radiological picture should be consistent with the timing of maternal infection before a genetic cause is excluded [8, 12].

8.4. Genetic phenocopies: recurrence, cascade testing and prenatal diagnosis

In the autosomal recessive phenocopies (AGS for most genes, USP18, OCLN, STAT2, ISG15, JAM3, RNASET2, NDE1, ANKLE2), the risk for each sibling is 25%, the parents are obligate carriers and unaffected siblings have a 2/3 probability of being carriers [4, 12]. In AGS there are autosomal dominant forms, due to specific heterozygous TREX1 and ADAR variants and to IFIH1 variants, de novo or inherited [4]. Once the familial variants have been identified, prenatal diagnosis and preimplantation genetic testing are available [12]. JAM3 cataracts may be seen on ultrasound at 23 weeks, but a normal fetal brain ultrasound does not exclude the disease [12]. For OCLN, family studies must use a method that detects copy number variants [12, 64].

8.5. Genetic deafness and cCMV

In early non-syndromic hereditary deafness, about 80% of cases are recessive [21]. The empirical recurrence risks for a sporadic case, without an environmental or syndromic diagnosis, are about 10% in the Oxford Desk Reference (some families with 25% and others with much less, indistinguishable) [21]. For hearing couples of Northern European ancestry with one deaf child and a negative family history, GeneReviews gives 18% for subsequent children, 14% if the child does not have GJB2 deafness and about 25% if the couple is consanguineous; two parents with GJB2 deafness have a 100% risk [5]. These figures do not apply if the deafness is attributed to confirmed cCMV, which gives CMV testing, in particular on the dried blood spot of children diagnosed after 3 weeks, direct value in counselling [5, 12]. Detection of GJB2 carriers is possible, and prenatal diagnosis of non-syndromic deafness is technically possible but controversial [21]. When one of the parents is profoundly deaf, the consultation should have a sign language interpreter available [21].

8.6. Counselling of children and adolescents

The sources consulted do not describe specific recommendations for counselling minors in cCMV. In girls with a history of cCMV, acquired seropositivity does not exempt them from hygiene measures in a future pregnancy [8]. In the recessive phenocopies, carrier testing of asymptomatic minor siblings has no immediate clinical utility in childhood and may be deferred until reproductive age, in line with general genetic counselling practice.

9. Special Considerations: Population Diversity and Epidemiology

9.1. Birth prevalence and what each figure measures

Maternal seroprevalence explains about 29–30% of the variability in birth prevalence between populations; other population factors are a high prevalence of HIV and a younger mean maternal age [3, 6]. Individual risk factors include young age, high parity, non-white race, low socioeconomic status, prematurity and NICU admission [3, 6]. The proportion that the CDC estimate to have birth defects or long-term health problems is about 1 in 5 NB with cCMV, and the European consensus gives 17–20% permanent sequelae in infected children [8, 30].

9.2. United States

IgG seroprevalence between 16 and 49 years is about 50%, ranging between 30 and 70% depending on geography and socioeconomic status; in NHANES data, about 1 in 60 seronegative people aged 12–49 years seroconvert each year, with a mean age at infection of 28.6 years, and IgM is positive in 2.3–4.5% of women of reproductive age, of whom only 14–18% have primary infection [3]. In the CHIMES study, infants identified by their mothers as black were about twice as likely to have cCMV as non-Hispanic white infants, after adjustment for maternal age and type of insurance [3]. In the KID database (2000–2012), NB with cCMV disease or isolated SNHL were hospitalised for 15–27 days versus 2 days, with costs of 45,771–89,846 dollars versus 2284–3813 dollars (2012 values), and about 50% were born before 37 weeks [3].

9.3. Europe

Seroprevalence in European pregnant women is 50–85%, and about half of cCMV results from maternal primary infection [8]. The risk of primary infection is higher in women who are young, multiparous, born in high-income countries and with a new pregnancy within 2 years of the previous one; in France, birth in a high-income country and high socioeconomic status were associated with cCMV after primary infection, and young age, birth in a low- or middle-income country and not working outside the home with cCMV after non-primary infection [3, 8]. In seropositive pregnant women, exposure to young children was not associated with non-primary infection or cCMV in Europe and Japan, unlike what was observed in Brazil [8]. The Dutch national study with retrospective diagnosis on dried blood spots and the French and German cohorts described in section 3 are the main European sources on natural history [3, 44, 45]. Seroprevalence is high in Japan, Taiwan and Sweden, and lower in the USA, Canada and Western Europe [3].

9.4. Portugal, Brazil and other regions

The main sources consulted do not contain estimates of the prevalence of cCMV in Portugal or Brazil, which does not demonstrate an absence of published data. From Brazil, the sources record the association between exposure to young children and non-primary infection or cCMV in seropositive women, and the prevalence of congenital Zika syndrome (3.8 per 10,000 live births in 2015 and 3.1 per 10,000 in 2016), relevant to the differential diagnosis [8, 67]. Australia contributes the cerebral palsy register, in which 9.6% of children had cCMV, and Canada the Ontario universal screening programme [3, 48].

9.5. Special populations

In very low birth weight NB (<1500 g), the incidence of cCMV is similar to that of the general population: in one centre, over 16 years, 18/4594 (0.39%; 95% CI 0.25–0.62) [3]. In preterm infants under 32 weeks, physical examination, full blood count, liver function tests and cranial ultrasound are already routine, MRI is performed if ultrasound is abnormal and is most informative at term-corrected age, and screening for retinopathy of prematurity (under 30 6/7 weeks or under 1500 g, at 4–9 weeks of life) may have to be brought forward in cCMV [3]. Saliva is less accurate in very preterm infants, and testing before 21 days makes it possible later to distinguish postnatal infection, to which very preterm infants are vulnerable [3, 28]. HIV-exposed NB are at increased risk of cCMV [3]. In pregnant women, working in day care and contact with young children increase the risk of primary infection [8].

9.6. Reference networks and patient organisations

There is no ERN dedicated to congenital infections. The networks relevant to the sequelae and to the differential diagnosis are ERN CRANIO, which includes rare ear, nose and throat disorders (https://www.ern-cranio.eu/), ERN-ITHACA for malformation syndromes and intellectual disability (https://ern-ithaca.eu/), ERN-RND for rare neurological diseases, including leukoencephalopathies such as AGS (https://www.ern-rnd.eu/), EpiCARE for rare epilepsies (https://www.epi-care.eu/) and ERN-RITA for rare immunological diseases, including the interferonopathies (https://ern-rita.org/). The European Congenital Infection Initiative (https://ecci.group/) publishes the European consensus, and the international CCMVNET registry (https://ccmvnet.org/) collects treatment data and runs a monthly virtual clinic [8]. Patient organisations include the National CMV Foundation in the USA (https://www.nationalcmv.org/), co-author of the position in favour of universal screening, and CMV Action in the United Kingdom (https://cmvaction.org.uk/) [10, 57]. For families of deaf children, the Oxford Desk Reference mentions in the United Kingdom the National Deaf Children's Society, Action on Hearing Loss and DeafPLUS [21]. The European consensus recommends informing families about local and national support organisations and about psychological support [8].

10. Illustrative Clinical Case

Vanbuggenhout and colleagues, from University Hospitals Leuven (Belgium), published in 2022 the case of a 29-year-old Caucasian woman, G3P1A1, a nursery school teacher, seronegative for CMV before pregnancy, with seroconversion in the 1st trimester [32]. Amniocentesis at 20 weeks had a positive CMV PCR. Fetal neurosonography was normal at 22 weeks; at 24 weeks it showed a bilateral hyperechogenic periventricular zone, confirmed at 26–27 weeks, with mild lenticulostriate vasculopathy. Fetal MRI at 27 weeks showed normal to slightly delayed gyration, small bilateral temporal cysts, hyperintensity at the crossroads near the basal ganglia and a band of hyperintensity above the lateral ventricles, with a normal apparent diffusion coefficient. At 30 weeks, neurosonography showed more severe lenticulostriate vasculopathy, bilateral temporal cysts and a slightly irregular ventricular contour, and MRI confirmed these findings with gyration appropriate for gestational age. Prenatal counselling predicted a risk of SNHL and of mild to moderate motor impairment, without severe cerebral palsy. No calcification was seen on prenatal imaging [32].

Labour was induced at 39+1 weeks because of arrested fetal growth and ended in vacuum delivery, because of fetal distress and meconium-stained fluid. The NB had an Apgar score of 9/9/10, weight of 2440 g (3rd centile), length of 44.5 cm (below the 3rd centile) and head circumference of 32 cm (3rd–10th centile). He presented with petechiae due to severe thrombocytopenia (13 × 10⁹/L) and hypoglycaemia, was transferred to a tertiary NICU and received platelet transfusions on days 1, 2, 4 and 7. CMV PCR on urine was strongly positive (>5,000,000 IU/mL), ABR suggested severe bilateral SNHL and fundoscopy showed no signs of CMV. Cranial ultrasound on day 1 showed germinolysis, temporal cysts and extensive bilateral calcifications, mainly periventricular, not seen prenatally (Figure 1 of the article: https://pmc.ncbi.nlm.nih.gov/articles/PMC9118604/figure/Fig1/). Oral valganciclovir was started on day 5, for a total of 6 months [32].

MRI on day 19 confirmed multiple periventricular and subependymal calcifications and showed delayed myelination of the posterior limb of the internal capsule, cerebellar hypoplasia, diffuse white matter abnormalities with increased diffusion and suspected polymicrogyria, as well as sequelae of intracranial haemorrhage attributed to the thrombocytopenia and vacuum delivery (Figure 2, comparison of prenatal T2 MRI with the postnatal susceptibility sequence: https://pmc.ncbi.nlm.nih.gov/articles/PMC9118604/figure/Fig2/; Figure 3, MRI on day 19: https://pmc.ncbi.nlm.nih.gov/articles/PMC9118604/figure/Fig3/). At 8 months, the child was receiving physiotherapy twice a week and had mild motor delay with mild limb hypertonia; the bilateral SNHL recovered, with near-normalisation on the left and persistent moderate SNHL on the right [32].

The authors draw two lessons. The first is that a 1st-trimester infection can progress to severe brain injury after 30 weeks, which is why they propose fetal neurosonography every 2 weeks and repeat fetal MRI at the end of the 3rd trimester when there are signs of active infection; this proposal is the authors' own and does not appear as a guideline recommendation, the guidelines advocating serial ultrasound and 3rd-trimester MRI without a fixed interval [8, 32]. The second is the relationship between the timing of infection and the neuroradiological pattern, summarised in Figure 4 of the article (https://pmc.ncbi.nlm.nih.gov/articles/PMC9118604/figure/Fig4/): neuronal proliferation between 8 and 16 weeks, migration between 12 and 20 and cortical organisation from 22, with lissencephaly, polymicrogyria, cerebellar hypoplasia and calcifications in early infections and white matter inflammation with cysts in late ones [32]. The case also shows the limits of prenatal imaging in detecting calcifications, the overlap between infectious and haemorrhagic lesions and the possibility of hearing improvement under treatment in an individual case, from which efficacy cannot be inferred. The management (valganciclovir started on day 5, for 6 months, in disease with CNS involvement) is consistent with the European and American recommendations [3, 8]. The figures of the article are licensed under CC BY 4.0.

11. Prognosis and Long-term Follow-up

11.1. Prognosis according to the timing of infection and fetal imaging

The strongest prognostic factor before birth is the gestational age at maternal primary infection: sequelae are concentrated in periconceptional and 1st-trimester infection (32.4% SNHL or neurological sequelae in the French cohort; 23% long-term sequelae in the European consensus) and are rare after infection in the 2nd and 3rd trimesters [8, 29, 45]. Based on serology alone, the SMFM estimates that a documented primary infection without confirmed fetal infection carries about a 3% risk of severe fetal sequelae, 8% of any adverse outcome and more than a 90% probability of an outcome without sequelae [43]. In the infected fetus, normal ultrasound and MRI give a negative predictive value close to 100% for moderate to severe sequelae, with a residual 17% risk of unilateral SNHL; isolated extracerebral findings are associated with a 30% risk and severe brain anomalies with a poor prognosis; the Cannie score grades the risk of SNHL from 0% (scores 1–2) to 66.7% (score 5) (section 5) [8].

11.2. Neonatal prognosis

Neuroimaging is the most reliable indicator of CNS involvement: normal imaging predicts normal or near-normal development, major lesions a poor prognosis, and isolated lenticulostriate vasculopathy or subtle white matter abnormalities are of uncertain significance [8]. In the Houston cohort of 41 children with symptomatic disease, microcephaly was the most specific predictor of intellectual impairment (100%; 95% CI 84.5–100) and of major motor disability (92.3%; 74.8–99), and abnormal CT the most sensitive; about 29% had an IQ or developmental quotient above 90, and SNHL at follow-up, but not SNHL at birth, was associated with a lower IQ [102]. The table in the Oxford Desk Reference, derived from this cohort, summarises the relationship [23].

Microcephaly relative to weight has high specificity for a poor neurological prognosis, unlike isolated symmetrical IUGR [8]. The series collected in the supplements to the European consensus identify microcephaly, abnormal CT or MRI and the severity of the neuroimaging score as predictors of poor development, and petechiae, IUGR and periventricular cysts on MRI as predictors of SNHL (De Cuyper, 2023; 1033 NB) [8]. The predictors on postnatal MRI described in the literature are cortical malformations, ventriculomegaly, hippocampal dysplasia, white matter abnormalities and calcifications, the latter associated with epilepsy [32]. Blood viral load is higher in symptomatic infants and in those who develop sequelae, and several studies associate low loads (for example, <10,000 copies/mL, NPV of 100%, or <1000 copies/mL, NPV of 100% in a series of 256 NB) with the absence of sequelae, but the thresholds vary between assays and there is no consensus [8]. Mortality in disease is 0.5–12% [3].

11.3. Long-term prognosis and effect of treatment

Lifelong disability affects 40–58% of NB with disease and about 13.5% of NB without signs, mainly through SNHL (section 3) [3, 7]. Children without signs up to 12–24 months have development equal to that of controls up to 18 years [3]. SNHL may progress and appear late: in the cohort without signs at birth, hearing abnormalities rose from 7% at 3 months to 14% at 5 years and 25% at 18 years, although the incidence between 6 and 18 years did not differ from that of controls [3]. Valganciclovir for 6 months increased the proportion of children with improved or normal hearing at 12 and 24 months (73% versus 57% and 77% versus 64%) and language outcomes at 24 months, with no effect on the primary outcome at 6 months [9]; intravenous ganciclovir for 6 weeks had improved hearing at 6 months, but the benefit waned over time, and in the Lanzieri cohort SNHL progressed with age in treated and untreated children [3, 9]. The natural history under treatment beyond 2 years therefore remains poorly defined.

11.4. Surveillance

In-office hearing screening is performed with otoacoustic emissions or automated ABR in both ears and, in cCMV disease, may be replaced by formal audiology [3]. US early hearing detection and intervention programmes cover 98% of NB, with diagnosis by 3 months and intervention by 6 months, but about 1 in 3 referred children is lost to follow-up [3]. Frequent follow-up in the first 2 years is justified because this is the period of greatest risk of SNHL and the critical period for language, and it should continue through childhood because hearing continues to deteriorate [28, 30]. The two guidelines diverge on ophthalmology and on extending hearing follow-up to low-risk children; the AAP applies the same hearing schedule to all children with cCMV, whereas the European consensus adjusts it according to the timing of maternal infection [3, 8].

12. Summary and Final Considerations

Congenital CMV infection affects about 0.64–0.7% of births in high-income countries and is the most common non-genetic cause of childhood SNHL [3, 6, 7]. Only 10–15% of infected NB have recognisable disease at birth; of these, 40–58% are left with permanent sequelae, and of NB without signs about 13.5% develop sequelae, mainly SNHL, which may be late-onset and progressive [3, 7, 8]. After maternal primary infection, transmission increases with gestational age (from 5.5% for preconceptional infection to 66.2% in the 3rd trimester) while sequelae are concentrated in periconceptional and 1st-trimester infection [29, 45]. Maternal immunity reduces transmission by about 69%, but not disease once the fetus is infected [3, 31].

Neonatal diagnosis requires PCR on urine or saliva collected in the first 3 weeks of life; after that period, only PCR on the newborn screening dried blood spot allows a retrospective diagnosis, with variable sensitivity that prevents a negative result from being used to exclude infection [3, 8, 63]. In pregnancy, IgG avidity dates the primary infection, and PCR on AF confirms fetal infection from 17 weeks and 8 weeks after primary infection according to the European consensus, or after 21 weeks and 6 weeks according to the SMFM [8, 43]. Oral valganciclovir 16 mg/kg twice daily for 6 months is the treatment for moderate to severe disease, off-label, with neutropenia as the main toxicity; 6-week treatment of isolated SNHL rests on non-randomised studies, and starting after the first month, in a randomised trial of older children, did not improve hearing [3, 8, 9, 89]. Valaciclovir 8 g/day in early primary infection reduced transmission (aOR 0.34 in the individual participant data meta-analysis) and is recommended by the European consensus, but not routinely in the USA; hyperimmune globulin failed in two randomised trials, the mRNA-1647 vaccine failed in phase 3 and prenatal letermovir is under trial [3, 27, 79, 84, 85, 88, 96]. Universal newborn screening has been implemented in regions of North America, the AAP recommends hearing-targeted screening with expanded testing and the UK NSC maintains the recommendation not to screen [3, 51, 56].

For the geneticist, cCMV has two roles. It is the diagnosis to exclude in the presence of intracranial calcifications, polymicrogyria, microcephaly, cystic leukoencephalopathy or SNHL, before attributing the picture to a monogenic disorder and counselling a 25% recurrence risk; and the genetic phenocopies, in particular the type I interferonopathies (AGS1–AGS9, USP18, STAT2, ISG15), OCLN, JAM3 and RNASET2, are the diagnosis to look for when the infection is not confirmed or does not explain the picture [4, 12, 19]. In deafness, a positive CMV result does not obviate genetic assessment, because the two causes may coexist and some genetic causes now have specific treatment [3, 42].

13. References

1. Orphanet. Fetal cytomegalovirus syndrome (ORPHA:294). https://www.orpha.net/en/disease/detail/294
2. Weller TH, Hanshaw JB. Virologic and clinical observations on cytomegalic inclusion disease. N Engl J Med. 1962;266:1233-1244. doi:10.1056/NEJM196206142662401.
3. Tesini BL, Caserta MT, Park AH, Puopolo KM, Reilly BK, Kuo DZ, et al.; Committee on Infectious Diseases, Section on Otolaryngology–Head and Neck Surgery, Committee on Fetus and Newborn, Council on Children With Disabilities. Care of the infant with congenital cytomegalovirus infection: clinical report. Pediatrics. 2026;158(4):e2026078595. doi:10.1542/peds.2026-078595.
4. Crow YJ. Aicardi-Goutières Syndrome. 2005 Jun 29 [updated 2016 Nov 22]. In: Adam MP, et al., editors. GeneReviews® [Internet]. Seattle (WA): University of Washington, Seattle; 1993–2026. https://www.ncbi.nlm.nih.gov/books/NBK1475/
5. Shearer AE, Hildebrand MS, Odell AM, Smith RJH. Genetic Hearing Loss Overview. 1999 Feb 14 [updated 2023 Apr 6; revised 2026 Jun 2]. In: Adam MP, et al., editors. GeneReviews® [Internet]. Seattle (WA): University of Washington, Seattle; 1993–2026. https://www.ncbi.nlm.nih.gov/books/NBK1434/
6. Kenneson A, Cannon MJ. Review and meta-analysis of the epidemiology of congenital cytomegalovirus (CMV) infection. Rev Med Virol. 2007;17(4):253-276. doi:10.1002/rmv.535.
7. Dollard SC, Grosse SD, Ross DS. New estimates of the prevalence of neurological and sensory sequelae and mortality associated with congenital cytomegalovirus infection. Rev Med Virol. 2007;17(5):355-363. doi:10.1002/rmv.544.
8. Leruez-Ville M, Chatzakis C, Lilleri D, Blazquez-Gamero D, Alarcon A, Bourgon N, et al. Consensus recommendation for prenatal, neonatal and postnatal management of congenital cytomegalovirus infection from the European congenital infection initiative (ECCI). Lancet Reg Health Eur. 2024;40:100892. doi:10.1016/j.lanepe.2024.100892. Corrigendum: Lancet Reg Health Eur. 2024;42:100974.
9. Kimberlin DW, Jester PM, Sánchez PJ, Ahmed A, Arav-Boger R, Michaels MG, et al. Valganciclovir for symptomatic congenital cytomegalovirus disease. N Engl J Med. 2015;372(10):933-943. doi:10.1056/NEJMoa1404599.
10. Pesch MH, Brown KD, Birath AL, Demmler-Harrison GJ, Sapp C, Selleck AM, et al. American Cochlear Implant Alliance position statement on newborn congenital cytomegalovirus screening. Otolaryngol Head Neck Surg. 2025;172(2):728-733. doi:10.1002/ohn.1079.
11. Orphanet. Pseudo-TORCH syndrome type 1 (ORPHA:1229). https://www.orpha.net/en/disease/detail/1229
12. GeneGuidance. Local curation of gene–disease validity and clinical notes for USP18, OCLN, JAM3, TREX1, RNASEH2A, RNASEH2B, RNASEH2C, SAMHD1, ADAR, IFIH1, STAT2, ISG15, RNASET2, NDE1 and ANKLE2 (extracted 17 Oct 2026), with the primary references cited by PMID.
13. Orphanet. Pseudo-TORCH syndrome type 2 (ORPHA:481665). https://www.orpha.net/en/disease/detail/481665
14. Orphanet. Pseudo-TORCH syndrome type 3 (ORPHA:714429). https://www.orpha.net/en/disease/detail/714429
15. Orphanet. Aicardi-Goutières syndrome (ORPHA:51). https://www.orpha.net/en/disease/detail/51
16. Uggenti C, Lepelley A, Depp M, Badrock AP, Rodero MP, El-Daher MT, et al. cGAS-mediated induction of type I interferon due to inborn errors of histone pre-mRNA processing. Nat Genet. 2020;52(12):1364-1372. doi:10.1038/s41588-020-00737-3.
17. Orphanet. Porencephaly-microcephaly-bilateral congenital cataract syndrome (ORPHA:306547). https://www.orpha.net/en/disease/detail/306547
18. Orphanet. Cystic leukoencephalopathy without megalencephaly (ORPHA:85136). https://www.orpha.net/en/disease/detail/85136
19. Firth HV, Hurst JA. Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. p. 200–202 (Intracranial calcification).
20. Smith RJH, Azaiez H, Booth K. GJB2-Related Autosomal Recessive Nonsyndromic Hearing Loss. 1998 Sep 28 [updated 2023 Jul 20]. In: Adam MP, et al., editors. GeneReviews® [Internet]. Seattle (WA): University of Washington, Seattle; 1993–2026. https://www.ncbi.nlm.nih.gov/books/NBK1272/
21. Firth HV, Hurst JA. Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. p. 122–125 (Deafness in early childhood).
22. Smith RJH, Azaiez H, Odell AM. SLC26A4-Related Sensorineural Hearing Loss. 1998 Sep 28 [updated 2025 Jan 9; revised 2025 Apr 3]. In: Adam MP, et al., editors. GeneReviews® [Internet]. Seattle (WA): University of Washington, Seattle; 1993–2026. https://www.ncbi.nlm.nih.gov/books/NBK1467/
23. Firth HV, Hurst JA. Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. p. 716–717 (Cytomegalovirus (CMV)).
24. Rawlinson WD, Boppana SB, Fowler KB, Kimberlin DW, Lazzarotto T, Alain S, et al. Congenital cytomegalovirus infection in pregnancy and the neonate: consensus recommendations for prevention, diagnosis, and therapy. Lancet Infect Dis. 2017;17(6):e177-e188. doi:10.1016/S1473-3099(17)30143-3.
25. Smith MG. Propagation in tissue cultures of a cytopathogenic virus from human salivary gland virus (SGV) disease. Proc Soc Exp Biol Med. 1956;92(2):424-430. doi:10.3181/00379727-92-22498.
26. Kimberlin DW, Lin CY, Sánchez PJ, Demmler GJ, Dankner W, Shelton M, et al. Effect of ganciclovir therapy on hearing in symptomatic congenital cytomegalovirus disease involving the central nervous system: a randomized, controlled trial. J Pediatr. 2003;143(1):16-25. doi:10.1016/S0022-3476(03)00192-6.
27. Shahar-Nissan K, Pardo J, Peled O, Krause I, Bilavsky E, Wiznitzer A, et al. Valaciclovir to prevent vertical transmission of cytomegalovirus after maternal primary infection during pregnancy: a randomised, double-blind, placebo-controlled trial. Lancet. 2020;396(10253):779-785. doi:10.1016/S0140-6736(20)31868-7.
28. Luck SE, Wieringa JW, Blázquez-Gamero D, Henneke P, Schuster K, Butler K, et al. Congenital cytomegalovirus: a European expert consensus statement on diagnosis and management. Pediatr Infect Dis J. 2017;36(12):1205-1213. doi:10.1097/INF.0000000000001763.
29. Chatzakis C, Ville Y, Makrydimas G, Dinas K, Zavlanos A, Sotiriadis A. Timing of primary maternal cytomegalovirus infection and rates of vertical transmission and fetal consequences. Am J Obstet Gynecol. 2020;223(6):870-883.e11. doi:10.1016/j.ajog.2020.05.038.
30. Centers for Disease Control and Prevention. Cytomegalovirus (CMV) and congenital CMV infection: clinical overview, laboratory testing, congenital CMV infection, congenital CMV and hearing loss [Internet]. Atlanta (GA): CDC; [accessed 17 Oct 2026]. Available from: https://www.cdc.gov/cytomegalovirus/hcp/clinical-overview/index.html
31. Fowler KB, Stagno S, Pass RF. Maternal immunity and prevention of congenital cytomegalovirus infection. JAMA. 2003;289(8):1008-1011. doi:10.1001/jama.289.8.1008.
32. Vanbuggenhout L, Aertsen M, De Catte L, Naulaers G. Pre- and postnatal brain magnetic resonance imaging in congenital cytomegalovirus infection: a case report and a review of the literature. BMC Pediatr. 2022;22(1):293. doi:10.1186/s12887-022-03334-x.
33. Ferreira CR, Darling A, Vockley J. Disorders of nucleic acid metabolism, tRNA metabolism and ribosomal biogenesis. In: Saudubray JM, Baumgartner MR, García-Cazorla Á, Walter J, editors. Inborn Metabolic Diseases: Diagnosis and Treatment. 7th ed. Berlin: Springer; 2022. Chapter 39, section 39.1.2.
34. Alsohime F, Martin-Fernandez M, Temsah MH, Alabdulhafid M, Le Voyer T, Alghamdi M, et al. JAK inhibitor therapy in a child with inherited USP18 deficiency. N Engl J Med. 2020;382(3):256-265. doi:10.1056/NEJMoa1905633.
35. Duncan CJA, Thompson BJ, Chen R, Rice GI, Gothe F, Young DF, et al. Severe type I interferonopathy and unrestrained interferon signaling due to a homozygous germline mutation in STAT2. Sci Immunol. 2019;4(42):eaav7501. doi:10.1126/sciimmunol.aav7501.
36. Martin-Fernandez M, Bravo García-Morato M, Gruber C, Murias Loza S, Malik MNH, Alsohime F, et al. Systemic type I IFN inflammation in human ISG15 deficiency leads to necrotizing skin lesions. Cell Rep. 2020;31(6):107633. doi:10.1016/j.celrep.2020.107633.
37. Calame DG, Wiener EK, Gavazzi F, Sevagamoorthy A, Pizzino A, Arnold K, et al. PTPN1-related autoinflammation is a common cause of Aicardi-Goutières syndrome with reduced penetrance [preprint]. medRxiv. 2026. doi:10.64898/2026.03.27.26345228.
38. O'Driscoll MC, Daly SB, Urquhart JE, Black GC, Pilz DT, Brockmann K, et al. Recessive mutations in the gene encoding the tight junction protein occludin cause band-like calcification with simplified gyration and polymicrogyria. Am J Hum Genet. 2010;87(3):354-364. doi:10.1016/j.ajhg.2010.07.012.
39. Mochida GH, Ganesh VS, Felie JM, Gleason D, Hill RS, Clapham KR, et al. A homozygous mutation in the tight-junction protein JAM3 causes hemorrhagic destruction of the brain, subependymal calcification, and congenital cataracts. Am J Hum Genet. 2010;87(6):882-889. doi:10.1016/j.ajhg.2010.10.026.
40. Freeman PJ, Hart RK, Gretton LJ, Brookes AJ, Dalgleish R. VariantValidator: accurate validation, mapping, and formatting of sequence variation descriptions. Hum Mutat. 2018;39(1):61-68. doi:10.1002/humu.23348. Tool: https://variantvalidator.org/
41. National Center for Biotechnology Information. ClinVar. VCV000009628 (m.1555A>G), VCV000009563 (m.7445A>G), VCV000042226 (m.7471dup), VCV000978789 (GJB6 deletion). Available from: https://www.ncbi.nlm.nih.gov/clinvar/
42. U.S. Food and Drug Administration. FDA approves first-ever gene therapy treatment for genetic hearing loss under national priority voucher [press release]. 2026 Apr 23. Available from: https://www.fda.gov/news-events/press-announcements/fda-approves-first-ever-gene-therapy-treatment-genetic-hearing-loss-under-national-priority-voucher
43. Society for Maternal-Fetal Medicine (SMFM); Hughes BL, Gyamfi-Bannerman C. Diagnosis and antenatal management of congenital cytomegalovirus infection. Am J Obstet Gynecol. 2016;214(6):B5-B11. doi:10.1016/j.ajog.2016.02.042.
44. Enders G, Daiminger A, Bäder U, Exler S, Enders M. Intrauterine transmission and clinical outcome of 248 pregnancies with primary cytomegalovirus infection in relation to gestational age. J Clin Virol. 2011;52(3):244-246. doi:10.1016/j.jcv.2011.07.005.
45. Faure-Bardon V, Magny JF, Parodi M, Couderc S, Garcia P, Maillotte AM, et al. Sequelae of congenital cytomegalovirus following maternal primary infections are limited to those acquired in the first trimester of pregnancy. Clin Infect Dis. 2019;69(9):1526-1532. doi:10.1093/cid/ciy1128.
46. Firth HV, Hurst JA. Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. p. 738–739 (Hyperechogenic bowel).
47. Tortora M, Fabbri E, Arossa A, Furione M, Chiapparini L, Conti G, et al. Serial prenatal MRI evaluation of valacyclovir-treated vs untreated fetuses with congenital cytomegalovirus (cCMV) infection. Eur Radiol. 2026. doi:10.1007/s00330-026-12698-y.
48. Leahy KE, Dunn JKE, Gallagher L, Reuvers E, Chakraborty P, Bitnun A, et al. Eye findings following congenital cytomegalovirus newborn screening. Pediatrics. 2026;158(4):e2025075586. doi:10.1542/peds.2025-075586.
49. Farrar DS, Knowles RL, Nunez C, Baynam G, Bitnun A, Duncanson M, et al. Etiology of severe microcephaly in infants: a multinational surveillance study. Pediatrics. 2026;157(2):e2025072700. doi:10.1542/peds.2025-072700.
50. Wats A, Araujo BL, Aparicio-Llorente C, Ganem J, Oliva IO, Gruen JR, et al. Hearing-targeted congenital cytomegalovirus screening: an 11-year experience. Pediatrics. 2026;158(2):e2025074229. doi:10.1542/peds.2025-074229.
51. Tavakoli NP, Sack V, Handel AS, Giacinto A, Pearce M, Ojukwu I, et al. Universal newborn screening for congenital cytomegalovirus using dried blood spot specimens. JAMA Netw Open. 2026;9(1):e2554518. doi:10.1001/jamanetworkopen.2025.54518.
52. Boppana SB, Ross SA, Shimamura M, Palmer AL, Ahmed A, Michaels MG, et al. Saliva polymerase-chain-reaction assay for cytomegalovirus screening in newborns. N Engl J Med. 2011;364(22):2111-2118. doi:10.1056/NEJMoa1006561.
53. Holden L, Carlson AM, Jayawardena A, Meyer A, Roby B, Lunos S, et al. The impact of mandated universal screening for congenital cytomegalovirus: the Minnesota experience. Otolaryngol Head Neck Surg. 2026;174(4):997-1002. doi:10.1002/ohn.70158.
54. Kunzweiler C, Stetson RC, Jacobson RM, St Sauver JL, Li Y, Jiang R, et al. Impact of hearing-targeted and universal newborn screening on the observed birth prevalence of congenital cytomegalovirus infection. J Pediatr. 2026;288:114811. doi:10.1016/j.jpeds.2025.114811.
55. Permar SR, Pesch MH, Schleiss MR. The need for universal congenital cytomegalovirus screening. Pediatrics. 2026;158:e2026076289. doi:10.1542/peds.2026-076289.
56. UK National Screening Committee. Screening for cytomegalovirus — cover sheet (newborn and antenatal screening review 2025). 2026 Jun 25. Available from: https://view-health-screening-recommendations.service.gov.uk/review/cytomegalovirus-newborn-and-antenatal-screening-review-2025/download-documents/cover_sheet/
57. UK National Screening Committee. Consultation opens on population screening for cytomegalovirus [blog]. 2025 Nov 14. Available from: https://nationalscreening.blog.gov.uk/2025/11/14/consultation-opens-on-population-screening-for-cytomegalovirus/
58. Fourgeaud J, Barry M, Bourgon N, Veyrenche N, Guilleminot T, Ville Y, et al. Cytomegalovirus vertical transmission rate according to IgG avidity value and valacyclovir treatment of maternal primary infection in the first trimester of pregnancy. Clin Infect Dis. 2025;81(5):e417-e423. doi:10.1093/cid/ciaf437.
59. Ornaghi S, Fernicola F, Carletti V, Corsi Decenti E, Dell'Oro S, Farsetti D, et al. Screening of cytomegalovirus infection in pregnant women. Recommendation by the Italian National Institute of Health and Ministry of Health, December 2023. Prenat Diagn. 2026. doi:10.1002/pd.70095.
60. Ornaghi S, Zammarchi L, Fernicola F, Tavanti M, de Chaurand VP, Boccalini S. Clinical and economic impact of universal screening for cytomegalovirus infection among pregnant women in Italy. Ultrasound Obstet Gynecol. 2026;67(6):764-773. doi:10.1002/uog.70235.
61. Raptopoulou A, Boureka E, Liberis A, Michos G, Kalogiannidis I, Mamopoulos A, et al. Cytomegalovirus infection in pregnancy: a comparative review of guidelines. Obstet Gynecol Surv. 2025;80(12):769-781. doi:10.1097/OGX.0000000000001450.
62. Boppana SB, Ross SA, Novak Z, Shimamura M, Tolan RW, Palmer AL, et al. Dried blood spot real-time polymerase chain reaction assays to screen newborns for congenital cytomegalovirus infection. JAMA. 2010;303(14):1375-1382. doi:10.1001/jama.2010.423.
63. Dollard SC, Dreon M, Hernandez-Alvarado N, Amin MM, Wong P, Lanzieri TM, et al. Sensitivity of dried blood spot testing for detection of congenital cytomegalovirus infection. JAMA Pediatr. 2021;175(3):e205441. doi:10.1001/jamapediatrics.2020.5441.
64. Jenkinson EM, Livingston JH, O'Driscoll MC, Desguerre I, Nabbout R, Boddaert N, et al. Comprehensive molecular screening strategy of OCLN in band-like calcification with simplified gyration and polymicrogyria. Clin Genet. 2018;93(2):228-234. doi:10.1111/cge.13025.
65. Li MM, Tayoun AA, DiStefano M, Pandya A, Rehm HL, Robin NH, et al. Clinical evaluation and etiologic diagnosis of hearing loss: a clinical practice resource of the American College of Medical Genetics and Genomics (ACMG). Genet Med. 2022;24(7):1392-1406. doi:10.1016/j.gim.2022.03.018.
66. Jones KL, Jones MC, del Campo M. Smith's Recognizable Patterns of Human Malformation. 8th ed. Philadelphia: Elsevier; 2022. p. 812–815 (Congenital Zika syndrome).
67. Sena LC, Ribeiro EM. Síndrome congénita pelo Zika vírus. In: Ribeiro EM, coord. Síndromes Genéticas: Reconhecer para Tratar. Fortaleza: EdUnichristus; 2021. p. 176–178.
68. Firth HV, Hurst JA. Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. p. 790–791 (Toxoplasmosis).
69. Firth HV, Hurst JA. Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. p. 782–783 (Rubella).
70. Meuwissen ME, Schot R, Buta S, Oudesluijs G, Tinschert S, Speer SD, et al. Human USP18 deficiency underlies type 1 interferonopathy leading to severe pseudo-TORCH syndrome. J Exp Med. 2016;213(7):1163-1174. doi:10.1084/jem.20151529.
71. Rasmussen SA, Jamieson DJ, Honein MA, Petersen LR. Zika virus and birth defects — reviewing the evidence for causality. N Engl J Med. 2016;374(20):1981-1987. doi:10.1056/NEJMsr1604338.
72. Spinner NB, Loomes KM, Krantz ID, Gilbert MA. Alagille Syndrome. 2000 May 19 [updated 2024 Jan 4]. In: Adam MP, et al., editors. GeneReviews® [Internet]. Seattle (WA): University of Washington, Seattle; 1993–2026. https://www.ncbi.nlm.nih.gov/books/NBK1273/
73. Firth HV, Hurst JA. Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. p. 208–211 (Leukodystrophy/leukoencephalopathy).
74. Firth HV, Hurst JA. Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. p. 228–231 (Microcephaly).
75. Cheplapharm/H2-Pharma. VALCYTE (valganciclovir) tablets and oral solution. Prescribing information. Revised 2025-12 (SPL effective 2025-12-08). DailyMed setid 4c517a39-2ded-4c5a-8d56-276853414b31. Available from: https://dailymed.nlm.nih.gov/dailymed/drugInfo.cfm?setid=4c517a39-2ded-4c5a-8d56-276853414b31
76. Neon Healthcare Ltd. Valcyte 450 mg film-coated tablets. Summary of Product Characteristics (UK). Electronic Medicines Compendium; last updated 2026-05-06. Available from: https://www.medicines.org.uk/emc/product/14226/smpc
77. GlaxoSmithKline. VALTREX (valacyclovir hydrochloride) caplets. Prescribing information (SPL effective 2025-11-12). Available from: https://dailymed.nlm.nih.gov/dailymed/search.cfm?query=valtrex
78. U.S. Food and Drug Administration. Drugs@FDA: CYTOVENE (ganciclovir) NDA019661. Available from: https://www.accessdata.fda.gov/scripts/cder/daf/index.cfm?event=overview.process&ApplNo=019661
79. Chatzakis C, Shahar-Nissan K, Faure-Bardon V, Picone O, Hadar E, Amir J, et al. The effect of valacyclovir on secondary prevention of congenital cytomegalovirus infection, following primary maternal infection acquired periconceptionally or in the first trimester of pregnancy. An individual patient data meta-analysis. Am J Obstet Gynecol. 2024;230(2):109-117.e2. doi:10.1016/j.ajog.2023.07.022.
80. Chatzakis C, Bourgon N, Fourgeaud J, Leruez-Ville M, Ville Y. The effect of valacyclovir on secondary prevention of congenital cytomegalovirus infection. Best Pract Res Clin Obstet Gynaecol. 2025;103:102679. doi:10.1016/j.bpobgyn.2025.102679.
81. Brafman J, Schlesinger Y, Barg M, Benenson S. Acute kidney injury, secondary to valacyclovir use during pregnancy for the prevention of fetal cytomegalovirus infection. Int J Gynaecol Obstet. 2026;175(1):493-494. doi:10.1002/ijgo.71021.
82. Leruez-Ville M, Ghout I, Bussières L, Stirnemann J, Magny JF, Couderc S, et al. In utero treatment of congenital cytomegalovirus infection with valacyclovir in a multicenter, open-label, phase II study. Am J Obstet Gynecol. 2016;215(4):462.e1-462.e10. doi:10.1016/j.ajog.2016.04.003.
83. Kagan KO, Bissinger AL, Roth O, Sonek J, Hoopmann M, Ville Y, et al. Predictive value of gB2 antibodies for maternal-fetal transmission after primary cytomegalovirus infection treated with valacyclovir. Ultrasound Obstet Gynecol. 2026;67(4):455-460. doi:10.1002/uog.70206.
84. Revello MG, Lazzarotto T, Guerra B, Spinillo A, Ferrazzi E, Kustermann A, et al. A randomized trial of hyperimmune globulin to prevent congenital cytomegalovirus. N Engl J Med. 2014;370(14):1316-1326. doi:10.1056/NEJMoa1310214.
85. Hughes BL, Clifton RG, Rouse DJ, Saade GR, Dinsmoor MJ, Reddy UM, et al. A trial of hyperimmune globulin to prevent congenital cytomegalovirus infection. N Engl J Med. 2021;385(5):436-444. doi:10.1056/NEJMoa1913569.
86. European Medicines Agency. Download medicine data: medicines output — medicines report. Accessed 2026-10-17. Available from: https://www.ema.europa.eu/en/medicines/download-medicine-data
87. Merck Sharp & Dohme. PREVYMIS (letermovir) tablets, oral pellets, injection. Prescribing information (SPL effective 2026-01-14). Available from: https://dailymed.nlm.nih.gov/dailymed/search.cfm?query=prevymis
88. Moderna, Inc. Moderna announces phase 3 study of investigational cytomegalovirus (CMV) vaccine did not meet primary efficacy endpoint [press release]. 2025 Oct 22. Available from: https://www.newswire.com/news/moderna-announces-phase-3-study-of-investigational-cytomegalovirus-cmv-vaccine
89. Kimberlin DW, Aban I, Peri K, Nishikawa JK, Bernatoniene J, Emonts M, et al. Oral valganciclovir initiated beyond 1 month of age as treatment of sensorineural hearing loss caused by congenital cytomegalovirus infection: a randomized clinical trial. J Pediatr. 2024;268:113934. doi:10.1016/j.jpeds.2024.113934.
90. Chung PK, Schornagel FAJ, Soede W, van Zwet EW, Kroes ACM, Oudesluys-Murphy AM, et al. Valganciclovir in infants with hearing loss and clinically inapparent congenital cytomegalovirus infection: a nonrandomized controlled trial. J Pediatr. 2024;268:113945. doi:10.1016/j.jpeds.2024.113945.
91. ClinicalTrials.gov. NCT03107871: Randomized controlled trial of valganciclovir for cytomegalovirus infected hearing impaired infants (ValEAR). Available from: https://clinicaltrials.gov/study/NCT03107871
92. ClinicalTrials.gov. NCT03301415: Asymptomatic congenital CMV treatment. NIAID. Available from: https://clinicaltrials.gov/study/NCT03301415
93. ClinicalTrials.gov. NCT05085366: A study to evaluate the efficacy, safety, and immunogenicity of mRNA-1647 cytomegalovirus (CMV) vaccine in healthy participants 16 to 40 years of age (CMVictory). Last update posted 2025-12-15. Available from: https://clinicaltrials.gov/study/NCT05085366
94. Fierro C, Brune D, Leggett R, Peterson J, Lorenz B, Calabro Calheiros R, et al. Safety and immunogenicity of the cytomegalovirus vaccine mRNA-1647 in healthy adults: results from a phase 2, randomized, controlled, dose-finding trial with long-term extension follow-up through month 48. Vaccines (Basel). 2026;14(5):444. doi:10.3390/vaccines14050444.
95. Faure Bardon V, Bouazza N, Peytavin G, Le MP, Benaboud S, Foissac F, et al. Maternal-fetal letermovir exposure in vivo and physiologically based pharmacokinetic modeling: a rationale for congenital cytomegalovirus therapy. J Infect Dis. 2026;233(5):848-857. doi:10.1093/infdis/jiaf619.
96. ClinicalTrials.gov. NCT05446571: Prenatal treatment of congenital cytomegalovirus infection with letermovir randomized against valaciclovir (step 2) (CYMEVAL III). Assistance Publique – Hôpitaux de Paris. Last update posted 2026-03-31. Available from: https://clinicaltrials.gov/study/NCT05446571
97. ClinicalTrials.gov. NCT06118515: A phase I pharmacokinetic and safety assessment of oral letermovir in infants with symptomatic congenital cytomegalovirus disease. NIAID. Available from: https://clinicaltrials.gov/study/NCT06118515
98. ClinicalTrials.gov. NCT07450365: Evaluation of long-term efficacy of 4 to 6-month course antiviral therapy for neurodevelopmental impairments caused by congenital cytomegalovirus infection. Available from: https://clinicaltrials.gov/study/NCT07450365
99. Sorrenti S, Elbarbary N, D'Antonio F, Di Mascio D, Khalil A. Diagnosis and management of congenital cytomegalovirus: critical appraisal of clinical practice guidelines. Eur J Obstet Gynecol Reprod Biol. 2025;306:172-180. doi:10.1016/j.ejogrb.2025.01.020.
100. Ito Y, Morioka I, Takahashi N, Fujioka K, Miura K, Moriuchi H, et al. Clinical practice guidelines for the management of congenital cytomegalovirus infection in Japan 2023: executive summary. Pediatr Infect Dis J. 2024;43(11):e390-e396. doi:10.1097/INF.0000000000004489.
101. Rybak-Krzyszkowska M, Stopyra L, Jaczynska R, Swiatkowska-Freund M, Borowski D, Czech-Kowalska J, et al. Recommendations of the Polish Society of Gynaecologists and Obstetricians and the Polish Society of Epidemiologists and Doctors of Infectious Diseases for screening, diagnosis and management of cytomegalovirus infection in pregnancy and congenital CMV infection. Ginekol Pol. 2025;96(10):857-876. doi:10.5603/gpl.108135.
102. Noyola DE, Demmler GJ, Nelson CT, Griesser C, Williamson WD, Atkins JT, et al. Early predictors of neurodevelopmental outcome in symptomatic congenital cytomegalovirus infection. J Pediatr. 2001;138(3):325-331. doi:10.1067/mpd.2001.112061.