Update on congenital cytomegalovirus infection: Prenatal prevention, newborn diagnosis, and management

---

The Canadian Paediatric Society gives permission to print single copies of this document from our website. For permission to reprint or reproduce multiple copies, please see our copyright policy.

Cytomegalovirus (CMV) is the most common cause of congenital infection, affecting 0.5% to 1% of live births in North America and Europe, and up to 6% of live births in developing countries ^[1]-[4]. CMV is the leading (up to 25%) cause of non-genetic childhood sensorineural hearing loss (SNHL) ^[5]. The estimated prevalence of long-term neurological sequelae ranges from 5% to 15% in asymptomatic congenital CMV (cCMV) cases, to 36% to 90% ^[1][6]-[10] in survivors of symptomatic cCMV (Figure 1).  

Recent advances in drug therapy for this infection have renewed focus on prevention, diagnosis, and treatment ^[5][11]. This practice point focuses on the diagnosis of cCMV in infants, and outlines current management strategies. Universal infant screening for cCMV is being considered in some countries, and in Canada, has already commenced in Ontario. 

Prenatal maternal infection and risk of fetal transmission

Expectant mothers with primary infection or non-primary (reinfection with different viral strains or reactivation of the primary strain) CMV are at risk of transmitting CMV vertically (transplacentally) to their unborn fetuses ^[12]-[15]. Rates of transmission are 10- to 15-fold higher in primary compared with non-primary maternal infection (Figure 1) ^{[1][6][10][16]}. However, most infected newborns are asymptomatic at birth, with only 10% demonstrating clinical features ^[5][9][11].  
 

Figure 1. Rates of fetal infection and sequelae following maternal infection.

Data from references [1][6][7][9][10][16]

CMV is transmitted horizontally through intimate contact with body fluids: saliva, urine, sexual secretions, and breast milk ^[17]. Among healthy young children in group care or crowded housing situations, opportunities for person-to-person transmission may be higher, thus potentially increasing the prevalence of primary infection. Infected children may continue to shed virus into their pre-school years ^[18] and are a major source of infection for their adult caregivers (including pregnant women) and other children. 

Although administering CMV-specific hyperimmune globulin and antiviral therapy to pregnant women with primary infection may provide benefit, robust evidence to support these interventions is lacking ^{[11][19]-[21]}. Developing an effective pre-pregnancy vaccine would likely be the best preventive strategy, but this option is not yet available ^[22]. Prospective mothers should be educated about hygienic measures, since these have been shown to decrease risk for maternal acquisition ^[23]-[25] (Table 1).

Table 1. Measures to reduce maternal risk of acquiring CMV infection
1. Avoid contact with body fluids
*A. Avoid mucosal to mucosal contact with saliva: Do not share toothbrushes Do not put infant pacifiers or objects child has mouthed in your mouth Do not share utensils, food, or drink used by young children Avoid contact with saliva when kissing children
B. Avoid new sexual partners: New partners who are CMV-positive create exposures through sexual secretions and saliva
2. Keep hands and surfaces clean*
†Hand hygiene after: Diaper change (or toilet use), nose-wiping Contact with toys, surfaces Handling child’s dirty laundry	‡Conduct regular cleaning of fomites: Clean diaper change area after each use Regularly clean or wash toys Regularly wipe surfaces that young children touch
*Measures recommended by Centers for Disease Control and Prevention, U.S. †Hand hygiene: Soap and water (15 to 20 seconds washing) or alcohol-based sanitizers are adequate ‡Effective surface cleaning agents for the home setting: soap and water, diluted bleach, or alcohol
Data from references [11][16][24][25]

Approach to investigating and evaluating infants at risk for cCMV

Clinical features of symptomatic cCMV at birth include microcephaly, intrauterine growth restriction (IUGR), hepatosplenomegaly, petechial rash, jaundice, and seizures (Table 2) ^[26]-[35]. Infants with symptomatic cCMV should be identified promptly, such that appropriate management can be instituted as early as possible to improve outcomes ^[36][37]. Diagnostic tests for cCMV rely on directly identifying the virus in an infant’s body fluids before 21 days of age ^[36][37]. When testing is delayed beyond 21 days, a positive result might indicate perinatal or postnatal infection, and is thus not conclusive for cCMV. Serology is not recommended because, although a negative IgG result can exclude cCMV, a positive result in infancy usually reflects passively transferred antibodies from an infected mother, making it unhelpful in confirming a diagnosis of cCMV ^[5][38]. Similarly, IgM performs poorly (sensitivity 20% to 70%) as a diagnostic test ^[39]. 

Table 2. Features of symptomatic* congenital CMV Infection in newborn infants

Physical examination findings

General: SGA, microcephaly†, jaundice, hydrops† Skin: Petechiae† Respiratory: Pneumonitis§ Abdomen: Hepatosplenomegaly† CNS: Seizures‡, Poor suck, hypotonia, lethargy  Hearing*  Eye*: Chorioretinitis†1, optic atrophy, microphthalmia, retinal scars, strabismus, cortical visual impairment‡

Laboratory abnormalities

Haematological: Low platelets2 most commonly but other cell line suppression may occur Transaminitis: Elevated ALT2 Hyperbilirubinemia: Increased conjugated bilirubin2 Spinal fluid abnormalities (not otherwise explained): Pleocytosis, elevated protein, positive CMV PCR‡

Head imaging abnormalities*1

Ventricular/Periventricular‡: calcifications† (50% of CNS abnormalities), ventriculomegaly ± atrophy Structural‡: Cerebellar/ependymal/parenchymal cysts Cortical‡: Polymicrogyria Migrational‡: Lissencephaly, porencephaly, schizencephaly, extensive encephalopathy Vascular: Lenticulostriate vasculopathy¶

1. Chorioretinitis and intracranial imaging abnormalities are considered highly supportive features of cCMV. 2. Non-transient elevations in transaminases (>2 times normal), bilirubin (>1 to 2 times normal), and platelets <100 x 109/L  without other explanation. *SNHL is only determined following audiological exam, which is usually not immediately available when symptom-based classification is performed in the initial newborn assessment; as such newborn hearing screen failure should trigger CMV testing if SNHL cannot be promptly ruled out. †The presence of these features or in conjunction with other findings (SNHL is included) are sufficient to trigger testing of infant’s urine or saliva for CMV. Single isolated non-highlighted features have a low positive predictive value. Some experts who favor universal screening feel that an isolated finding is sufficient to trigger testing ‡ Clinically significant CNS disease § Although rare and more likely to occur with preterm or immunodeficient hosts with postnatal CMV, sepsis-like presentations, pneumonitis, and colitis have been reported in cCMV in association with prematurity and congenital immunodeficiency.  ¶ Common finding of questionable clinical significance. Some experts believe its presence in isolation is not sufficient to classify infant as moderate to severe. ALT Alanine transaminase; CMV Cytomegalovirus; cCMV Congenital CMV; CNS Central nervous system; PCR Polymerase chain reaction; SNHL Sensorineural hearing loss; SGA Small for gestational age

Data from references [5] [26]-[35]

Testing using CMV rapid viral cultures (sensitivity >92%, specificity 100%) or PCR-based assays (sensitivity >95%, specificity >98%) of the infant’s urine (preferable method and a bag specimen is adequate) or saliva should be done ^[40]-[42]. Saliva specimens should be collected by swabbing the inside of an infant’s mouth. Testing should be done more than 1 hour after breastfeeding because shorter intervals may increase false positive results in infants of CMV-infected mothers ^[41]. All positive saliva specimens should be confirmed with a urine test. In populations where CMV status is unknown but includes children who may not have clinical features of cCMV, the sensitivity of the newborn dried blood spot (DBS) has a wide range (28% to 73%) ^[43][44]. However, the sensitivity (of 84%) was higher in a meta-analysis that included infants with symptoms compatible with cCMV or those who had already been confirmed to have cCMV ^[45]. Infants are not necessarily viremic at birth. Therefore, a negative CMV polymerase chain reaction (PCR) test of whole blood or plasma, or of newborn DBS, does not exclude CMV ^[43][44]. A positive DBS result is useful, however, for confirming viremia ^[43][44][46], and may sometimes be helpful for evaluations that are delayed beyond postnatal age of 3 weeks.

Infants born to mothers with suspected or proven CMV during pregnancy (based on maternal symptoms, serology, fetal ultrasonographic features, or placental histopathology), or infants with risk factors (Table 3) or clinical features compatible with cCMV, should undergo CMV testing and clinical and laboratory evaluation to determine disease classification. Laboratory testing should evaluate for anemia, thrombocytopenia, hyperbilirubinemia, and transaminitis ^[26]. A negative urine or saliva CMV test should trigger investigations for other congenital infectious or genetic causes (Figure 2). 

Table 3. Identification, diagnosis, and evaluation of infants with cCM

What are the indications for infant testing? Antenatal Indicators of risk: Maternal CMV infection Fetal ultrasound with findings suggestive of cCMV Placental pathology consistent with CMV infection  Other indicators of potential risk: HIV exposure Primary immunodeficiency Features consistent with symptomatic cCMV (Table 2) A failed newborn hearing screen or confirmed SNHL How is cCMV diagnosis confirmed? ‘Gold standard’ test: Urine CMV PCR/shell vial before 21 days postnatal age Other positive tests before 3 days postnatal age: CMV PCR on a) newborn DBS, b) saliva PCR, if confirmed by urine or other*  How should infants be evaluated when cCMV has been confirmed? Blood: CBC, differential; Bilirubin; ALT, AST‡ ±CSF if done as part of workup for seizures/sepsis, request CMV PCR testing Imaging: HUS unless there are neurological concerns, then MRI† If HUS is abnormal, follow with MRI† Hearing evaluation Ophthalmological evaluation

*Blood or CSF are not recommended as routine tests, but if positive before 3 weeks post-birth, would confirm cCMV. †MRI has the advantage of being able to detect more of the cortical and hippocampal changes that predict longer neurological outcome when compared with CT scan. Due to radiation exposure, CT is not the preferred option. ‡Because AST elevations may be due to hemolysis, ALT elevations are more reliable. ALT Alanine transaminase; AST Aspartate transaminase; CBC Complete blood count; cCMV Congenital cytomegalovirus; CSF Cerebrospinal fluid; DBS Dried blood spot; HUS Head ultrasound; MRI Magnetic resonance imaging; PCR Polymerase chain reaction

 

Figure 2. Algorithm outlining testing, referral and treatment of infants suspected to have cCMV is available as a supplementary file.

Newborns who fail their newborn hearing screen (NBHS) are also being tested for CMV. This more targeted approach allows for earlier detection and management of infants with CMV-associated SNHL ^[30][47], and has been mandated in some U.S. states ^[48]. However, targeted screening will not identify those infants who are asymptomatic at birth and pass their newborn hearing assessment but go on to develop late-onset isolated SNHL.

Although universal screening would (ideally) optimize detection of cCMV, consensus is needed concerning the optimal specimen to use for early newborn screening. The range of sensitivity for DBS testing and the additional cost and logistical challenges of using better performing specimens, such as urine or saliva, are unresolved considerations ^[49]. 

If infants with suspicion for cCMV with CNS involvement have a lumbar puncture performed to exclude other infectious causes, CMV PCR testing on cerebrospinal fluid (CSF) can be considered ^[5]. A positive PCR in CSF offers diagnostic support for CNS disease (Table 3). However, the low sensitivity of PCR and its lack of utility as a prognostic measure make it difficult to recommend routine PCR testing for cases of symptomatic cCMV ^[50]. When testing has confirmed cCMV, symptomatic infants require ophthalmological and hearing evaluation as well as head imaging (Table 3) ^[5][11]. For asymptomatic or minimally symptomatic infants, a head ultrasound (HUS) is usually sufficient. Both MRI and CT scan neuroimaging are superior to ultrasound (US) for identifying abnormalities that predict less favourable outcomes ^[51]. However, MRI is preferred over CT and should be performed in infants with significant neurological features or in those with abnormal  findings on US (Table 3), because MRI has the added advantage of being better able to detect findings that predict neurodevelopmental morbidity (e.g., dysplasia of hippocampus and cerebellum, and polymicrogyria) (Table 3) ^[52].

Management approach based on symptom severity 

Cases of cCMV are classified at or around the time of birth by symptom severity as asymptomatic, mild, and moderate to severe disease (Table 4) ^[11][53]. Because the timing of symptom assessment and severity staging precedes audiological assessment, the currently accepted symptom-based classification makes no reference to the presence of SNHL. 

Table 4. Essential points for disease classification, referral, treatment recommendations, and follow-up

How are infants classified? Confirmed:  Positive ’gold standard’ test* before 3 weeks OR newborn DBS-positive Asymptomatic ± SNHL‡ No clinical or laboratory abnormality consistent with cCMV Mildly symptomatic ± SNHL 1 to 2 isolated, transient, mild features of cCMV without chorioretinitis or CNS involvement (e.g., mild transaminitis or transient thrombocytopenia) Moderate-to-severely symptomatic ± SNHL CNS disease: Microcephaly, seizures, positive CSF CMV PCR, abnormal head imaging Chorioretinitis Multisystem disease (≥3 organs/systems involved) with significant non-transient abnormal laboratory values (e.g., hepatosplenomegaly, IUGR and moderate-to-severe hepatitis and moderate-to-severe or persistent thrombocytopenia) Severe single organ disease† Probable§: Positive ‘gold standard’ test* after 3 weeks PLUS CMV-specific features (Table 2)   Who should be referred to an infectious disease specialist? All confirmed symptomatic cases All cases of cCMV with SNHL, even if asymptomatic Probable cases may be considered Who should be treated? CNS disease Chorioretinitis Severe single or multi-organ disease What is the recommended treatment? Antiviral treatment should commence in first month of life and be administered for 6 months Valganciclovir (16 mg/kg/dose by mouth, twice per day, for 6 months) Ganciclovir (6 mg/kg/dose IV twice per day) may be used for the first 2 to 6 weeks before transitioning to valganciclovir for very sick neonates What does follow-up entail? Monitoring on treatment CBC/differential weekly x 1 month; every 2 weeks x 2 months then monthly x 3 months AST, ALT, urea, creatinine monthly x 6 months Follow-up Audiology: Frequent tests in the first 2 to 3 years, then yearly until school age.   Testing frequency is determined by audiology based on results from previous testing. Ophthalmology: Evaluate as soon as possible after cCMV is confirmed. Follow-up testing is based on initial findings. Neurodevelopmental follow-up: Child should be followed closely in the first 2 years. Frequency of follow-up in first 2 years and thereafter should be determined by neurological concerns and initial assessment.

*‘Gold standard’ test refers to CMV PCR/CMV viral culture of urine OR saliva CMV PCR in non-breastfed infants. A positive saliva CMV PCR test requires confirmatory urine CMV testing in breastfed infants. †For example, severe hepatitis, severe bone marrow suppression (profound/ persistent/refractory pancytopenia not explained by another cause), colitis, pneumonitis (the latter two are rare but can be severe in preterm infants) (53) ‡Experts differ in treatment approach to moderately symptomatic infants and infants with SNHL. In absence of evidence, some experts consider treatment in isolated SNHL if it is identified before 1 month of age §May benefit from CMV PCR testing of DBS. A negative test does not rule out cCMV ALT Alanine transaminase; AST Aspartate transaminase; CNS Central nervous system; CBC Complete blood count; cCMV Congenital cytomegalovirus; DBS Dried  blood spot; PCR Polymerase chain reaction; SNHL sensorineural hearing loss

Asymptomatic disease 

There is no evidence to support antiviral therapy for asymptomatic CMV disease without SNHL at the present time. However, asymptomatic infants require regular audiological evaluation, with prompt ear, nose and throat (ENT) specialist referral if SNHL is detected. Expert opinions differ regarding treatment of infected infants with isolated SNHL ^[11][37]. Definitive recommendations await results of ongoing trial, but existing observational data suggest benefit ^[54].

Mild disease  

Mild CMV is characterized by transient or minor abnormalities in one or two organ systems, with no CNS involvement. Managing mild CMV cases who are later confirmed to have SNHL should involve ID consultation ^[11][37]. Management may be individualized, taking into consideration the presence and severity of SNHL ^[11][27] (Figure 2).

Moderate to severe disease 

All moderate to severe CMV cases (defined in Table 4) ^[53] should be referred promptly for infectious disease consultation. Data from two trials have provided support for treating severely symptomatic infants with antiviral agents ^[55][56]. The more recent trial demonstrated improved neurodevelopmental and hearing outcomes for infants treated for 6 months as compared with for 6 weeks ^[56]. Oral valganciclovir should be administered for 6 months. Parenteral ganciclovir may be substituted for the first 2 to 6 weeks of treatment when the infant is very ill ^[5][57]. 

Adverse effects (AEs) of antiviral therapy include neutropenia ^[29], thrombocytopenia, transaminitis, and elevated urea and creatinine levels. While on therapy, infants need to be monitored serially for AEs (Table 4). There is no evidence to support routine viral load monitoring, as viral loads do not consistently correlate with treatment response ^[28][56]. 

Follow-up and monitoring for sequelae (Table 4)

In addition to ongoing hearing and neurodevelopmental evaluation and ophthalmological follow-up ^[2][58], close dental follow-up may also be needed. Enamel hypoplasia has been reported in up to 40% of children with symptomatic CMV ^[59][60]. Evidence for a high prevalence of vestibular dysfunction in children with cCMV, even those without SNHL, is growing, which indicates that occupational therapy may also be of benefit ^[61]. 

Late-onset, progressive SNHL has a median onset of 27 months of age, but has been reported to develop as late as 44 months of age ^[30]. Serial hearing evaluations should be conducted regularly over the first 4 to 5 years of life ^[30]. Children with extensive CNS involvement may experience seizures, cerebral palsy, and intellectual delay (Table 5) ^{[17][62]-[64]}. They require neurodevelopmental follow-up and appropriate multidisciplinary support.
 

Table 5. Neurodevelopmental, hearing, and visual sequelae in infants with symptomatic cCMV
	Frequency (%) in infants symptomatic at birth
Any sequelae	36 to 90
Microcephaly	35 to 40
Severe motor deficits	15 to 25
Neurodevelopmental	42
Cognitive deficits	50 to 70
Seizures	15 to 20
Any SNHL	27 to 69
Unilateral SNHL	19
Bilateral SNHL	50
Ocular abnormalities	25 to 50
Visual impairment	22 to 58
Data from references [1][6][7][9][10][16][31][62]-[64]

Conclusion

cCMV is the leading cause of non-genetic SNHL and a significant cause of neurodevelopmental and neurosensory morbidity. Basic hygienic measures are highly effective in preventing maternal infection and should be promoted in antenatal care. Early laboratory identification of infants with cCMV (within the first 21 days after birth) is essential to determine disease severity and initiate antiviral therapy in moderate to severe cases. All infants with symptomatic cCMV and asymptomatic infants with isolated hearing loss should be referred to an infectious disease specialist. When indicated, valganciclovir therapy, initiated in the neonatal period and administered for 6 months, has been shown to improve hearing and developmental outcomes. Affected infants require multidisciplinary follow-up. Future research should address newborn screening, treatment outcomes in children with isolated SNHL or mild disease, CMV vaccine development, as well as diagnostic and prognostic strategies.

Acknowledgements

This practice point was reviewed by Dr. Soren Gantt (University of British Columbia, B.C. Children’s Hospital Research Institute) and Dr. Wendy Vaudry (University of Alberta, Stollery Children’s Hospital), with our thanks. It has also been reviewed by the Community Paediatrics, Drug Therapy and Hazardous Substances, Fetus and Newborn, and Mental Health and Developmental Disabilities Committees of the Canadian Paediatric Society.

---

CANADIAN PAEDIATRIC SOCIETY INFECTIOUS DISEASES AND IMMUNIZATION COMMITTEE
Members: Michelle Barton MD; A. Michael Forrester MD (past member); Ruth Grimes MD (Board Representative); Nicole Le Saux MD (Chair); Laura Sauve MD; Karina Top MD
Liaisons: Tobey Audcent MD, Committee to Advise on Tropical Medicine and Travel (CATMAT), Public Health Agency of Canada; Sean Bitnun MD; Canadian Pediatric and Perinatal HIV/AIDS Research Group; Fahamia Koudra MD, College of Family Physicians of Canada; Marc Lebel MD, IMPACT (Immunization Monitoring Program, ACTIVE); Yvonne Maldonado MD, Committee on Infectious Diseases, American Academy of Pediatrics; Jane McDonald MD, Association of Medical Microbiology and Infectious Disease Canada; Dorothy L. Moore MD, National Advisory Committee on Immunization (NACI); Howard Njoo MD, Public Health Agency of Canada
Consultant: Noni E MacDonald MD
Principal authors: Michelle Barton MD, A. Michael Forrester MD, Jane McDonald MD

---

References

---

Disclaimer: The recommendations in this position statement do not indicate an exclusive course of treatment or procedure to be followed. Variations, taking into account individual circumstances, may be appropriate. Internet addresses are current at time of publication.