Vous êtes sur la page 1sur 9

ESPID Reports and Reviews

Congenital Cytomegalovirus

Editor: Delane Shingadia
Board Members
David Burgner (Melbourne, Cristiana Nascimento-Carvalho George Syrogiannopoulos
Australia) (Bahia, Brazil) (Larissa, Greece)
Kow-Tong Chen (Tainan,Taiwan) Ville Peltola (Turku, Finland) Tobias Tenenbaum (Mannhein, Germany)
Luisa Galli (Florence, Italy) Emmanuel Roilides (Thessaloniki, Marc Tebruegge (Southampton, UK)
Steve Graham (Melbourne, Greece) Marceline Tutu van Furth (Amsterdam,
Australia) Ira Shah (Mumbai, India) The Netherlands)

Congenital Cytomegalovirus
A European Expert Consensus Statement on Diagnosis and Management
Suzanne E. Luck, MBChB, MD,*†‡ Jantien W. Wieringa, MD,§ Daniel Blázquez-Gamero, MD, PhD,¶
Philipp Henneke, MD,∥ Katharina Schuster, MD,∥ Karina Butler, MB, BCh, FRCPI,** Maria Grazia Capretti, MD, PhD,††
Maria José Cilleruelo, PhD,‡‡ Nigel Curtis, MA, MBBS, DCH, DTM&H, MRCP, FRCPCH, PhD,§§
Francesca Garofoli, PhD,¶¶ Paul Heath, MB BS, FRCPCH,* Elias Iosifidis, MD, MSc, PhD,∥∥
Nigel Klein, BSc, MBBS, PhD,*** Giuseppina Lombardi, MD,††† Hermione Lyall, BSc Hons, MBChB, MD, FRCPCH,‡‡‡
Tea Nieminen, MD, PhD,§§§ Dasja Pajkrt, MD, PhD, MBA,¶¶¶ Vassiliki Papaevangelou, MD, PhD,∥∥∥
Klara Posfay-Barbe, MD, MS,**** Laura Puhakka, MD,§§§ Emmanuel Roilides, MD, PhD, FIDSA, FAAM,∥∥
Pablo Rojo, MD PhD,¶ Jesús Saavedra-Lozano, MD, PhD,†††† Teshri Shah, MSc,‡‡‡ Mike Sharland, FRCPCH,*
Harri Saxen, MD, PhD,§§§ and Ann C.T.M Vossen, MD, PhD,‡‡‡‡ on Behalf of the ESPID Congenital
CMV Group Meeting, Leipzig 2015

Key Words: congenital CMV, investigation, in the developed world. Reported prevalence mately 25% of all children with sensorineural
management, treatment varies between cohorts but is approximately hearing loss (SNHL).1,2 Meta-analysis shows
7 per 1000 births.1 About half of cytomegalo- that although long-term sequelae, especially
virus (CMV)-infected babies with clinically SNHL, are more common in those with clini-
detectable disease at birth are destined to have cally detectable disease at birth, they are also

C ongenital cytomegalovirus (cCMV) is

the most common congenital infection
significant impairments in their development,
and cCMV infection is implicated in approxi-
found in 13% of those without clinical features
attributable to CMV on initial examination.1
Accepted for publication August 31, 2017. UCD School of Medicine and Health Sciences, Dub- §§Pediatric Infectious Diseases, Children’s Hos-
From the *Paediatric Infectious Diseases Research lin, Ireland; ††Department of Obstetrical, Gynae- pital, University of Helsinki and Helsinki Univer-
Group, St George’s University, London, United cological and Paediatric Sciences, Operative Unit sity Hospital, Finland; ¶¶¶Department of Pediatric
Kingdom; †Centre for Virology, University Col- of Neonatology, Polyclinic St. Orsola-Malpighi, Infectious Diseases, Emma Children’s Hospital,
lege Medical School, London; ‡Kingston Hospital University of Bologna, Bologna, Italy; ‡‡Pediatric Academic Medical Center, Amsterdam, The Neth-
NHS Foundation Trust, London, United Kingdom; Infectious Diseases, Hospital Universitario Puerta erlands; ∥∥∥National and Kapodistrian University of
§The Hague Medical Center (HMC), Department of de Hierro Majadahonda, Madrid, Spain; §§Depart- Athens, Greece; ****Children’s Hospital of Geneva,
Pediatrics and Sophia Children’s hospital, Erasmus ment of Paediatrics, The University of Melbourne University Hospitals of Geneva, Switzerland;
Medical Center Rotterdam Department of Pediatric ††††Pediatric Infectious Diseases Unit, Gregorio
& Murdoch Children’s Research Institute, Royal
Infectious Diseases; ¶Pediatric Infectious Diseases Marañón Hospital, Madrid, Spain; and ‡‡‡‡Depart-
Unit, Hospital Universitario 12 de Octubre, Uni- Children’s Hospital Melbourne, Parkville, Australia;
¶¶NICU and Neonatal Unit, Fondazione IRCCS ment of Medical Microbiology, Leiden University
versidad Complutense, Instituto de Investigación Medical Center, Leiden, The Netherlands.
Hospital 12 de Octubre, Madrid, Spain; ∥Center for Policlinico S. Matteo, Pavia, Italy; ∥∥Infectious Dis-
eases Unit, 3rd Department of Pediatrics, Faculty Dr Luck has recently provided consultancy services
Chronic Immunodeficiency and Center for Pediat- to Roche.
rics and Adolescent Medicine, Medical Center and of Medicine, Aristotle University School of Health
Sciences, Thessaloniki, Greece; ***Great Ormond The authors have no funding or conflicts of interest
Faculty of Medicine, University of Freiburg, Ger- to disclose.
many; **Our Lady’s Children’s Hospital Crumlin, Street Children’s Hospital, London, and the Institute
of Child Health, UCL, London, United Kingdom; Address for correspondence: Suzanne E. Luck,
Copyright © 2017 Wolters Kluwer Health, Inc. All rights MBChB, MD, Paediatric Infectious Diseases
reserved. †††NICU and Neonatal Unit, Fondazione IRCCS
Research Group, St George’s University, Jenner
ISSN: 0891-3668/17/3612-1205 Policlinico S. Matteo, Pavia, Italy; ‡‡‡Imperial Col-
Wing, Level 2, Room 2.216F, Mail Point J2C, Lon-
DOI: 10.1097/INF.0000000000001763 lege NHS Healthcare, London, United Kingdom;
don, SW17 0RE. E-mail: Suzanne.luck@ucl.ac.uk.

The ESPID Reports and Reviews of Pediatric Infectious Disease Journal series topics, authors and contents are chosen and
approved independently by the Editorial Board of ESPID.

The Pediatric Infectious Disease Journal  •  Volume 36, Number 12, December 2017 www.pidj.com | 1205

Copyright © 2017 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
Luck et al The Pediatric Infectious Disease Journal  •  Volume 36, Number 12, December 2017

Despite the significant long-term

TABLE 1.  Grade System of Evaluating Evidence3
impact of cCMV infection, there is limited
evidence on which to base many treatment Quality Rating Definition Example Methodology Depiction in Text
decisions in clinical practice. In an era of
enhanced perinatal screening, fetuses and High Further research is very Randomized trials or A
newborns are increasingly tested for CMV unlikely to change our double-upgraded
confidence in the estimate ­observational studies
after abnormalities were detected during rou- of effect
tine ultrasonography or maternal serology. Moderate Further research is likely to Downgraded randomized B
Furthermore, otherwise “asymptomatic”, have an i­ mportant impact trials or upgraded
congenitally CMV-infected, newborns are on our ­confidence in the observational studies
­estimate of effect and may
being identified after detection of SNHL change the estimate
through newborn hearing screening pro- Low Further research is very Double-downgraded C
grams. Because of earlier diagnosis, babies likely to have an important randomized trials or
with cCMV now presenting to pediatricians impact on our confidence observational studies.
in the ­estimate of effect
differ from those primarily included in clini- and is likely to change the
cal trials of treatment reported in the litera- estimate
ture. Very low Any estimate of effect is very Triple-downgraded D
A symposium was convened during uncertain ­randomized ­trials,
the 2015 conference of the European Soci- or downgraded
­observational studies, or
ety of Paediatric Infectious Diseases to dis- case series/case reports
cuss the current management of cCMV. In
Strength of Definition Depiction in Text
attendance were clinicians from throughout ­Recommendation
Europe, many of whom are involved in policy Strong Most informed patients would choose the ­recommended 1
for cCMV for their region/country. ­recommendation management and clinicians can structure their interac-
This article summarizes the discus- for using (or tions with patients ­accordingly
sions at this meeting alongside the evidence not using) an
informing them. A balanced perspective of
Weak Patients’ choices will vary according to their values and 2
the controversies in this area is presented ­recommendation preferences and clinicians must ensure that patients’
and areas of consensus highlighted. Finally, for using (or care is in keeping with their values and preferences
where evidence is lacking, suggestions are not using) an
made for future research efforts to address ­intervention
areas of unmet medical need. Strength of recommendations is determined by the balance between desirable and undesirable consequences of alter-
The authors acknowledge the coexist- native management strategies, quality of evidence, variability in values and preferences and resource use.

ing need for studies on the management of

babies with symptoms consistent with cCMV,
but in whom this diagnosis cannot be firmly TABLE 2.  Possible Signs and Symptoms in Children With Congenital CMV5–8
established, and of those with symptomatic
Clinically detectable symptoms/signs
postnatal CMV infection; this article does  Physical Examination
not, however, address these groups.   Small for gestational age (birth weight <−2 SD for gestational age)
The internationally accepted GRADE   Microcephaly (head circumference <−2 SD for gestational age)
system for evaluating evidence has been used   Petechiae or purpura (usually found within hours of birth and persist for several weeks)
to illustrate points where relevant (Table 1).3   Blueberry muffin rash (intra dermal hematopoiesis)
  Neurologic physical examination
DISEASE    Microcephaly (head circumference <−2 SD for gestational age)
Classically, cCMV infection is catego-    Neurologic signs (lethargy, hypotonia, seizures, poor sucking reflex)
rized as “symptomatic” or “asymptomatic” at Abnormalities detected incidentally or through subsequent investigation/specialist examination
birth. Differing definitions and opinions on  Laboratory results
what constitutes “symptomatic” CMV infec-
  Thrombocytopenia (occurs in the first week but platelets often increase spontaneously after
tion, however, makes interpreting the litera- the second week)
ture challenging. Indeed, some of the largest   Leukopenia, isolated neutropenia
cohort studies include babies with SNHL at   Elevated liver enzymes (ALT/AST)
birth in the group described as being “asymp-    Conjugated hyperbilirubinemia
 Cerebrospinal fluid
tomatic” because no “clinically apparent
  Abnormal cerebral fluid indices, positive CMV DNA
disease” was detectable during newborn  Neuroimaging
examination.4 In modern healthcare systems,   Calcifications, periventricular cysts, ventricular dilatation, subependymal pseudocysts, ger-
whereby cCMV is increasingly detected minolytic cysts, white matter abnormalities, cortical atrophy, migration disorders, cerebellar
through screening for other conditions, hypoplasia, lenticulostriatal vasculopathy
 Hearing test
alongside increased accessibility of investi-   Sensorineural hearing loss uni- or bilaterally
gations, such as magnetic resonance imaging  Visual examination
(MRI), the traditional dichotomy between   Chorioretinitis, retinal hemorrhage, optic atrophy, strabismus, cataracts
clinically “apparent” and “inapparent” dis- *CMV-associated jaundice can be present at the first day after birth and usually persists longer than physi-
ease is becoming less meaningful. Table 2 ologic jaundice.
summarizes the accepted clinical features of ALT indicates alanine aminotransferase; AST, aspartate aminotransferase; SD, standard deviations.

1206 | www.pidj.com © 2017 Wolters Kluwer Health, Inc. All rights reserved.

Copyright © 2017 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
The Pediatric Infectious Disease Journal  •  Volume 36, Number 12, December 2017 Congenital CMV: European Consensus

cCMV disease with those symptoms detecta- multiorgan involvement. Babies with addressed this specific population, but a
ble on newborn examination listed separately transient or otherwise clinically insig- nonrandomized cohort study observing
to those detectable only if specific investi- nificant abnormalities (ie, the babies the effects of valganciclovir in isolated
gations are conducted, for example, when are not “sick”) that resolve sponta- SNHL is in progress (clinicaltrials.gov
cCMV is already suspected.5–8 neously over a few weeks are not NCT02005822). The majority of experts
included in this group even if these at this meeting would categorize babies
Full Consensus Within This Expert abnormalities are multiple. with isolated, confirmed SNHL in the
Group Was That 2c. 
A further group exists that may be “severe”/CNS group because bilateral
considered to have “moderate” dis- SNHL is not only associated with likely
1. For the purposes of research and publica- ease. This group is heterogeneous long-term impairments but was also
tion, newborns identified as having cCMV and includes, for example, those with included in the criteria for recruitment
disease after abnormal clinical examina- persistent (eg, more than 2 weeks in the only randomized controlled trials
tion at birth (such as microcephaly, small duration) abnormalities of hemato- (RCTs) in cCMV. However, consensus
for gestational age (SGA), widespread logic/biochemical indices or more was not reached because the spectrum
petechiae, hepatosplenomegaly) should than 2 “mild” disease manifestations of hearing loss is wide, and treatment of
be differentiated from those babies iden- (as listed earlier). Because of lack of isolated SNHL has not been evaluated
tified through screening or investigation evidence, full consensus could not be in any RCTs.
for other disorders, for example, those reached on how to approach this group,
tested for CMV after known/likely mater- and treatment decisions are currently
nal infection or abnormal newborn hear- made on a case by case basis. Devel- WHEN SHOULD TESTING
ing screening. This differentiation would opment of a validated clinical scoring
allow for more accurate assessment of the system for disease severity at presen-
prognostic value of individual manifesta- tation and risk of sequelae would be CONSIDERED?
tions of “symptomatic” disease on longer- beneficial for both counseling parents Indications for testing for cCMV are
term outcomes as already shown in other and informing treatment decisions. based on the presence of one or more of the
publications.9 most frequently observed clinical features
2. “Symptomatic” cCMV should be consid- (Table 3).17 Unfortunately, predictive values
ered as “severe,” “moderate” or “mild” 3. Defining CNS involvement for each of these features are not available.
disease. a. It remains uncertain whether some,
a. “Mild” disease includes those with nonspecific findings detected on cra- Full Consensus Within This Expert
isolated (1 or 2 at most), otherwise, nial ultrasound (CrUSS) and MRI Group Was That Testing for
clinically insignificant or transient (particularly isolated lenticulostriatal cCMV Should be Performed in
findings, such as petechiae, mild vasculopathy [LSV]) constitute clini-
hepatomegaly or splenomegaly or bio- cally significant CNS disease. LSV 1. Fetuses with ultrasound/MRI imaging
chemical/hematologic abnormalities has been detected in 0.4%–5.8% of all consistent with cCMV disease (by appro-
(such as thrombocytopenia, anemia, neonates undergoing an ultrasound, priately timed antenatal testing of amni-
leukopenia, borderline raised liver and only 5% has been associated otic fluid).18 (Quality C, Level 1)
enzyme abnormalities or conjugated with cCMV.11,12 Some have suggested 2. Newborns where there is a maternal his-
hyperbilirubinemia) or SGA (defined isolated LSV as a marker of risk for tory of suspected primary CMV infection
as weight for gestational age <−2 SNHL.11 Others have found only during pregnancy. If antenatal testing of
standard deviations) without micro- more extensive neuroimaging abnor- amniotic fluid has been conducted, it is
cephaly. malities to be of prognostic value.13,14 suggested that cCMV infection should
b. “Severe” disease includes those The majority at this meeting would still be confirmed at birth because both
with central nervous system (CNS) not consider LSV in isolation to be a false-positive and -negative results have
involvement (abnormal neurologic or notable CNS manifestation of disease. been reported.18 (Quality C, Level 1)
ophthalmologic examination, micro- It is suggested that neuroradiologic 3. Newborns with signs/symptoms con-
cephaly or neuroimaging consistent abnormalities not known to be clearly sistent with cCMV disease (see Table 2;
with cCMV disease [such as calcifica- associated with CMV disease and including those with findings consistent
tions, moderate to severe ventriculo- adverse outcomes are discussed with with cCMV on antenatal imaging). (Qual-
megaly, cysts, white matter changes, a suitably experienced neuroradiolo- ity B, Strength 1)
cerebral or cerebellar hypoplasia, hip- gist, particularly, if the results of these 4. Children with confirmed SNHL.16 Sys-
pocampal dysplasia, neuronal migra- discussions might influence treatment tems need to be established to ensure
tion abnormalities])10 or with life- decisions. testing for cCMV occurs, where possible,
threatening disease. b. The exact pathophysiology of SNHL in the first 21 days of life because dried
is not clear but is likely secondary to blood spot (DBS) are not always readily
infection and degradation of sensory available for testing (see below). (Quality
The Majority Agreed That structures within the inner ear.15,16 It B, Strength 1)
is therefore debated whether isolated
2b. “Severe” disease also includes babies SNHL should truly be considered a The Majority Agreed That
with evidence of severe single-organ CNS manifestation of infection and,
disease (including those with clinically as a consequence, whether such chil- 5. Newborns who are SGA should not rou-
significant liver enzyme abnormalities dren should be considered comparable tinely be tested. Studies in SGA newborns
[liver “failure”] and marked hepatos- to those with CNS disease included in have shown the prevalence of cCMV to be
plenomegaly) or those with significant published clinical trials. No studies have 0%–5.2%.19–22 However, the majority of

© 2017 Wolters Kluwer Health, Inc. All rights reserved. www.pidj.com | 1207

Copyright © 2017 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
Luck et al The Pediatric Infectious Disease Journal  •  Volume 36, Number 12, December 2017

infection, and repeat sampling is not neces-

TABLE 3.  Clinical Features That Should Lead to Testing for Congenital CMV
sary. After 21 days, a urine positive for CMV
Neonates could be because of CMV acquired postna-
 Physical examination tally from, for example, passage through the
   Hepatosplenomegaly birth canal or through breast milk. As CMV
  Petechiae, purpura or blueberry muffin rash in a newborn PCR techniques are becoming more sensi-
  Jaundice (prolonged or conjugated hyperbilirubinemia)
tive, earlier testing, before the age of 14 days,
  Microcephaly (head circumference <−2 SD for gestational age)
  Consider if symmetrically small for gestational age (<−2 SD for gestational age) is recommended.27
 Neurology CMV PCR testing of saliva is an
  Seizures with no other explanation alternative and is easy to perform. Samples
 Laboratory parameters should be taken immediately before feeding
  Prolonged jaundice with transaminitis
   Conjugated hyperbilirubinemia
in breastfed newborns, and confirmed with
   Unexplained thrombocytopenia, consider if leucopenia or anemia urine, as false-positive results have been
 Neuroimaging reported.28–31
  Intracranial calcification (often periventricular) PCR assay of neonatal DBS can be
  Intracranial ventriculomegaly without other explanation performed retrospectively in an attempt to
  Consider in the case of periventricular cysts, subependymal pseudocysts, germinolytic cysts,
white matter abnormalities, cortical atrophy, migration disorders, cerebellar hypoplasia,
diagnose cCMV after the first 21 days of life.
lenticulostriate vasculopathy Sensitivity is around 84% in meta-analysis
 Visual examination but is highly variable depending on the labo-
  Abnormal findings on ophthalmologic examination consistent with congenital ratory techniques used and the population
CMV (eg, chorioretinitis)
being tested; a negative DBS PCR cannot,
  Consider if congenital cataracts
 Failed neonatal hearing screen therefore, be used to definitively exclude a
 Maternal serology diagnosis of cCMV.32
  Evidence of maternal seroconversion*
  Consider in women with known CMV infection (known IgG seropositive at start of preg- Full Consensus Within This Expert
nancy), particularly, if symptoms or virologic examination consistent with suspected CMV
Group Was That
Older children 1. Testing for cCMV should be performed
 Sensorineural hearing loss: new diagnosis using a single CMV PCR of urine obtained
Features in bold are those where there is consensus for testing. Features in italics are those that might lead to within 21 days of birth but ideally within
testing in individual circumstances and depending on local practice. 14 days of birth (Quality B, Strength 1).
*Seek expert clinical virology advice for interpretation of virologic investigations in pregnancy. 2. Saliva PCR testing can be an alternative,
†Baseline screening to differentiate between congenital and postnatal CMV infection is helpful for extremely pre-
mature infants (<28 weeks gestational age) who are at increased risk of symptomatic postnatal infection.
but a positive result should be confirmed
SD indicates SD indicates standard deviations. using urine (Quality B, Strength 1).
3. After the age of 21 days, CMV DNA PCR
of stored DBS can be used to diagnose
studies report a prevalence of 1.4%–1.8%, was not reached regarding practice in this cCMV retrospectively; sensitivity is rela-
which is not significantly higher than the area, with cost being a factor among other tively low, and a negative test cannot be
prevalence of cCMV in the general popu- considerations.24 (Quality C, Strength 2) used to definitively exclude a diagnosis of
lation. Therefore, evidence is insufficient 7. Testing of babies born to mothers who cCMV (Quality B, Strength 1).
to justify screening all newborns with are known to be CMV seropositive at the
isolated SGA for cCMV. None of these establishment of pregnancy. Although
studies distinguish between asymmetrical maternal nonprimary CMV infection is RECOMMENDED
(with normal head circumference) and known to be important when considering INVESTIGATIONS AFTER
symmetrical SGA, but when head circum- the overall burden of cCMV disease, test- CONFIRMING A DIAGNOSIS OF
ference was mentioned, most SGA babies ing all babies born to these women, par- CONGENITAL CMV INFECTION
with cCMV had microcephaly (head cir- ticularly in populations with high maternal
cumference <−2 standard deviations).21,22 After a virologic diagnosis of cCMV
seroprevalence, is tantamount to universal
Because of this, and the poor prognos- infection has been made, additional investi-
neonatal screening.25,26 Identifying women
tic outcome of children with cCMV and gations are necessary to evaluate the extent of
with nonprimary CMV who are at highest
microcephaly, many present at this meet- disease and to assist with discussions regard-
risk of transmitting infection to their fetus ing prognosis and treatment.
ing test those babies with symmetrical remains elusive. It was agreed that individ-
SGA but not those with preserved head ual case discussion and local policy should
growth.14(Quality C, Strength 2)
Full Consensus Within This Expert
therefore dictate practice in this area. Fur- Group Was That
6. Prematurity. Evidence that premature ther research is clearly needed.
babies have a higher incidence of cCMV
is limited.20,23 Testing extremely premature 1. The investigations below are conducted in
babies (<28 weeks gestational age) at birth LABORATORY DIAGNOSIS OF any baby in whom a diagnosis of cCMV
does, however, assist in differentiating CONGENITAL CMV INFECTION is confirmed, looking specifically for the
between congenital and postnatal infec- Testing for cCMV using CMV poly- manifestations of disease (Table 2):
tion. This may be very helpful in guiding merase chain reaction (PCR) in urine is • Complete blood count, liver enzymes,
the management of these babies that are highly reliable: sensitivity is 100% and speci- (conjugated) bilirubin
particularly vulnerable to symptomatic ficity 99%.27 One negative urine specimen in • Renal function (before initiating
postnatal infection. However, consensus a neonate is therefore sufficient to exclude ­therapy)

1208 | www.pidj.com © 2017 Wolters Kluwer Health, Inc. All rights reserved.

Copyright © 2017 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
The Pediatric Infectious Disease Journal  •  Volume 36, Number 12, December 2017 Congenital CMV: European Consensus

• CrUSS. (Quality A, Strength 1) of CMV disease at birth, some argued Although clinical findings such as SGA
• Audiologic testing (brainstem-evoked that it is desirable to conduct MRI in all and petechiae have been shown in histori-
response; some screening tests such cCMV-infected babies because additional cal cohorts to predict risk for SNHL, more
as otoacoustic emissions are not pathology can be identified as compared recent reanalysis of data indicates that these
sufficient to detect central auditory with CrUSS38–40 (Quality D, Strength 2). findings in isolation are generally associated
hearing loss in cCMV). (Quality A, with disease-free outcomes in babies present-
Strength 1) ing without other manifestations of sympto-
• Ophthalmic assessment. (Quality A,
No antiviral drugs are currently matic disease.9,46 Opinion on the severity,
Strength 1) or number, of symptoms justifying antiviral
2. If additional imaging to CrUSS is felt to licensed for the treatment of cCMV. Although
many case reports and cohort studies have treatment remains divided, and it is therefore
be indicated, then MRI is the preferred strongly recommended that clinicians dis-
neuroimaging modality. MRI can be suc- reported on treatment for cCMV, there are
results from only 2 RCTs.7,41–44 The first of cuss treatment initiation and duration with an
cessfully performed in neonates without expert in this area.
the need for sedation and is, therefore, these studies evaluated 6 weeks’ intrave-
both highly sensitive and free of the risks nous ganciclovir treatment in neonates (<1
month of age), gestational age ≥32 weeks Full Consensus Within This Expert
of radiologic exposure, which accom- Group Was That
pany computed tomography. (Quality C, and clinically apparent disease in the new-
Strength 1) born period with evidence of CNS disease
(including microcephaly, intracranial calcifi- 1. Babies with evidence of CNS disease
3. MRI should be performed in babies with
cation, abnormal CSF indices for age, hear- should receive antiviral treatment (Qual-
clinically detectable neurologic findings
ing deficit and chorioretinitis).7 Improved ity A, Strength 1). Treatment should be
or CrUSS abnormalities.
hearing and neurodevelopmental outcomes preferably for 6-months duration (Quality
were shown, but there was significant loss to B, Strength 2).
The Majority Agreed That follow-up.7,44 A more recent trial compared 2. Babies with no clinical/laboratory find-
6-week to 6-month treatment with oral val- ings consistent with CMV disease should
4. Cranial MRI should be performed in any ganciclovir and included babies with any not receive treatment because no evidence
babies with cCMV and evidence of CMV evidence of symptomatic (including non- exists to support treatment in this group
disease (see Table 2). (Quality C, Strength CNS) cCMV disease.41 Few babies enrolled, (Quality D, Strength 1 [not to treat]).
1) however, had isolated, mild clinical features, 3. Babies with evidence of life-threatening
5. CMV PCR quantitation should be per- and none in the 6-month treatment group had disease or severe single-organ disease or
formed in blood at baseline. Several isolated SNHL (D Kimberlin 2015, personal multiorgan involvement should receive
studies have shown the absence of CMV email correspondence, 28 April). A mod- treatment. Although evidence is limited,
viremia to be associated with better long- est benefit on both 2-year hearing and neu- particularly for life-threatening disease,
term outcomes, and this may be reassuring rodevelopmental outcomes was shown with consensus was that treatment should be con-
when evaluating babies without any other the 6-month treatment course. The longer sidered in this group (Quality B, Strength
manifestations of cCMV disease.33–35 treatment course improved likelihood of bet- 1). Consensus could not be reached on
Blood CMV PCR should not, however, be ter hearing outcomes most notably in those duration of treatment in this group.
used to rule out cCMV infection because, with preexisting CNS involvement. Longer 4. Oral valganciclovir is now the drug of
paradoxically, the absence of CMV in duration of therapy was only statistically choice. Intravenous ganciclovir should
blood has been described even in babies significant, however, for “total ear” hearing be used in babies unable to tolerate oral
with severe cCMV disease.33,36 (Quality C, as opposed to “best ear” hearing (which is
drug or where gastrointestinal absorption
Strength 2) of greater functional significance) and only
is uncertain (Quality A, Strength 1).
6. Examination of cerebrospinal fluid (CSF): once adjusted for baseline CNS involvement.
No current evidence supports examination Given the natural resolution of some fea- The Majority Agreed That
of CSF as part of routine diagnostic work tures of cCMV disease in published cohorts,
up. Studies have shown detectable CMV alongside the delayed onset of hearing loss
5. Babies with “mild” cCMV disease (as
DNA in CSF, and elevated biomarkers and fluctuations in SNHL reported in cCMV,
defined earlier) should not receive treat-
such as β2-microglobulin suggest a poor it is even more challenging to draw any con-
ment. No studies have clearly addressed
prognosis.13,37 However, others have shown clusions regarding treatment effect from
treatment in this group. Most present at
no additional prognostic value from CSF uncontrolled studies.7,16,45
Clinical trials to date do not, there- this meeting would not, therefore, treat
specimens obtained in the clinical set-
fore, provide good evidence on which to base babies with 1 or 2 isolated or transient,
ting.37 Despite this lack of evidence, there
treatment decisions for many of the infants clinically insignificant, manifestations of
was a majority view that although a pos-
sible area of interest for future research, presenting to clinicians in everyday clinical disease (Quality C, Strength 2).
lumbar puncture should not be performed practice. 6. Babies with “moderate” cCMV disease
routinely in babies with cCMV infection Table 4 provides guidance on which (as defined earlier). Evidence for treating
(Quality C, Strength 1). infants should be offered treatment after a babies with multiple, but not severe, mani-
risk versus benefit discussion with the family. festations of disease (including jaundice,
Only a Minority Agreed That This table and associated text indicate areas hepatosplenomegaly without significantly
where consensus was reached. Much discus- raised liver enzymes, SGA) is limited. It is,
7. Cranial MRI should be performed in all sion focused around the treatment of babies therefore, recommended that these cases are
CMV-infected babies. Although there is with less severe cCMV disease and whether discussed on a case-by-case basis with a cli-
no conclusive evidence that performing the minimal additional benefit shown in the nician with experience of managing babies
MRI gives additional prognostic informa- 6-month treatment course was sufficient to with cCMV (such as a pediatric infectious
tion to CrUSS in those without evidence justify such a prolonged course of treatment. disease specialist) (Quality B, Strength 2).

© 2017 Wolters Kluwer Health, Inc. All rights reserved. www.pidj.com | 1209

Copyright © 2017 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
Luck et al The Pediatric Infectious Disease Journal  •  Volume 36, Number 12, December 2017

TABLE 4.  Summary of Treatment Recommendations

Disease Manifestation Treatment Recommendation Level of Evidence

 CNS disease Ganciclovir/valganciclovir: duration
  Microcephaly, CNS calcification, chorioretinitis 6 months* Treatment: Quality A, Strength 1 (to treat)
  White matter changes (or other abnormalities on Duration: Quality B, Strength 2
MRI consistent with CMV disease)†
 Other “severe” disease (includes life-threatening or Ganciclovir/valganciclovir: minimum of Treatment: Quality B, Strength 1
  severe single-organ or multiorgan non-CNS 6 weeks, up to 6 months*‡ Duration: Quality B, Strength 2
 “Mild” disease: isolated or transient disease (eg, No treatment Treatment: Quality C, Strength 2 (for no
  jaundice, Petechiae, SGA in isolation; max 2 ­treatment)
 No clinical or biochemical findings of disease No treatment Treatment: Quality D, Strength 1 (for no
   (± detectable CMV viremia) ­treatment)
Majority opinion: but not consensus
 Isolated hearing deficit*§ Ganciclovir/valganciclovir: Duration Treatment: Quality C, Strength 1
6 months* Duration: Quality C, Strength 2
 “Moderate” disease (see text for definition; eg, multiple Consider treatment after discussion Treatment: Quality C, Strength 2
minor findings consistent with CMV disease)* with specialist
Duration: Minimum of 6 weeks and Duration: Quality B, Strength 2
up to 6 months*
There is currently only evidence for starting treatment in the first month of life.
*Limited evidence without full consensus: see text for further description.
†In the case of isolated, nonspecific MRI findings that are not consistent with cCMV disease, it was agreed that treatment is not necessarily indicated.
‡It was suggested (without consensus) that treatment might continue in this group until the underlying clinical manifestation of disease (eg, hepatitis) resolved because benefit of 6
months treatment is unclear.
§No studies address this particular group, although they were included in eligibility criteria for treatment in both published RCTs of treatment.

7. Treatment of isolated SNHL: The major- 10. Treating babies older than 28 days: Treat- compared with placebo in the only RCT evalu-
ity at this meeting would include SNHL ment of older children has not been ating this.41 The oral administration of valgan-
at birth in their indications for treatment addressed in any RCTs, although it is ciclovir also removes the burden of hospitali-
because this was in the inclusion criteria acknowledged that the 28-day cutoff is zation and risk of nosocomial infections and
for treatment in previous RCTs. Further- also not evidence based. Retrospective central line complications observed during
more, the main benefit of treatment is in case series of small numbers of babies treatment with ganciclovir. Hepatotoxicity has
preserving hearing rather than improving treated outside the newborn period have been reported in up to 30% of those treated with
hearing once damage exists, with good reported good outcomes.48,49 Babies ganciclovir and thrombocytopenia in a similar
outcomes reported in observational stud- found to have SNHL after hearing screen- proportion.51 In the most recent study of treat-
ies (with likely bias).7,41,47 There was not, ing at birth often do not have a diagno- ment with valganciclovir, deranged liver func-
however, consensus, and it is acknowl- sis of cCMV confirmed until outside the tion was observed, but this was neither clinically
edged that no RCTs have specifically 1-month “window of evidence” for treat- nor statistically significant when compared with
addressed treatment effect in this group ment. No consensus was reached on how placebo. In all studies, abnormal biochemical
of babies who are usually now identified late it might be acceptable to start treat- and hematologic parameters resolved after drug
through newborn hearing screening pro- ment in this scenario, or in the eventuality discontinuation.
grams (Grade C, Strength 1). of hearing deterioration. Two RCTs are Long-term side effects have not been
8. Drug dose and formulation: Although oral currently evaluating the use of treatment evaluated in neonates treated with ganciclo-
valganciclovir is now first-line treatment in older children with cCMV and SNHL vir or valganciclovir. Animal studies raise the
in most cases, it is currently unknown (clinicaltrials.gov NCT01649869 and theoretical risk of gonadotoxicity and car-
whether valganciclovir reaches target areas NCT02606266), which may clarify this cinogenicity.52,53 Although this has not been
as effectively as ganciclovir or, indeed, debate. (Evidence for treating outside the observed in humans to date, parents should
where drug should be targeted (eg, CNS or newborn period Quality D, Strength 2.) be counseled about these potential risks, par-
inner ear) because no studies have directly ticularly when considering treatment in those
compared the 2 drugs. In those with severe groups in which benefit has not been clearly
disease, particularly if absorption is uncer- SIDE EFFECTS OF ANTIVIRAL
shown. No adverse long-term effects have been
tain, intravenous ganciclovir is, therefore, TREATMENT documented in a small cohort of babies treated
preferred by some in early stages of treat- Much of the debate around treating less in early neonatal studies and followed up to
ment until oral therapy can be reliably tol- severely affected babies relates to the potential puberty (NCT00031421, unpublished data).
erated (Quality C, Strength 1). side effects of currently available antiviral drugs.
9. Treatment duration in cases without CNS Significant neutropenia is frequently
involvement: In those infants in whom observed during antiviral treatment in infants. MONITORING OF BABIES DURING
the decision is taken to give antiviral This is reported less commonly with valgan- TREATMENT
treatment, the majority would treat for 6 ciclovir than with ganciclovir (21% compared Table 5 summarizes a proposed moni-
months. However, there was no consensus with 65%).7,41,44,50 Neutropenia generally occurs toring strategy for babies treated for cCMV.
on this point in light of the modest benefit during the first month of treatment, with no These recommendations are based on the
shown for longer treatment courses in the increased toxicity observed after 6 weeks in safety monitoring and data obtained from the
only RCT (Quality B, Strength 2). those randomized to receive 6-month treatment published RCTs.7,41

1210 | www.pidj.com © 2017 Wolters Kluwer Health, Inc. All rights reserved.

Copyright © 2017 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
The Pediatric Infectious Disease Journal  •  Volume 36, Number 12, December 2017 Congenital CMV: European Consensus

TABLE 5.  Monitoring and Follow-Up According to Treatment Status

No Treatment Given Treatment Given

— Investigations whilst on treatment*

— FBC,* LFT† and U&E suggested weekly for first 4 weeks and then at least monthly
until completion of treatment course (ganciclovir/valganciclovir)‡ (Quality B,
Strength 2)
Weight measurement and drug dose review at time of blood sampling
— Viral load at baseline (Quality C, Strength 2).
Consider Viral load 2–4 weekly whilst on antiviral therapy (not consensus; Quality D,
Strength 2)§
— Consider therapeutic drug monitoring if:
 Viral load increase >1.0 log10 during treatment¶
 Toxicity is suspected
 There is an increased risk of toxicity: eg, prematurity <36 weeks, abnormal renal
 (Quality D, Strength 2)
Follow up Follow up
Audiology assessment every 3–6 months in the first year, then every 6 months until 3 years of age and then every 12 months until 6 years old∥
(Quality C, Strength 1)
Pediatric infectious disease clinic review (or general pediatric Pediatric infectious disease clinic as soon as possible in the first month, then annual
clinic after consultation with a specialist) until at least 1 review until at least age 2 years (specialist or general clinic with pediatric infectious
year, and ideally 2 years, of life. (Quality D, Strength 1) diseases input depending on local agreements). (Quality D, Strength 1)
Monitor development. (Quality D, Strength 1) Monitor development with neurodevelopmental assessment at 1 year in a child develop-
ment service. (Quality D, Strength 1)
Ophthalmic assessment as directed by ophthalmologist, but Ophthalmic assessment directed by ophthalmologist, but baseline and annual review
baseline and annual review up to age 5 years in those up to age 5 years recommended.** (Quality D, Strength 2)
with clinically detectable symptoms/signs at birth recom-
mended.** (Quality D, Strength 2)
FBC indicates full blood count; LFT, liver function tests; U&E, urea, creatinine and electrolytes.
*Interrupt treatment or consider granulocyte colony stimulating factor (GCSF) if absolute neutrophil count <0.5  × 109/L. Decreasing dose may be considered for less
severe neutropenia.
†LFT monitoring monthly is sufficient if sampling difficulties.
‡Increase frequency or seek advice if there is deterioration.
§Measuring viral load is not evidence based but offers some evaluation of virus response and enables detection of possible viral resistance.
¶Consider CMV resistance testing (sequencing) in unexplained elevations/breakthrough of viremia.
∥According to current United Kingdom newborn hearing screening guidelines.
**There is limited evidence on late ocular manifestations of cCMV. They are rare and include visual impairment and strabismus.6,53

There are no data to support therapeu- Only a Minority Agreed That for development of cCMV-associated hearing
tic drug monitoring.54 Therapeutic drug mon- loss and a critical period for language devel-
itoring may, however, have a role when toxic- 1. Viral load monitoring: Some centers opment. Early detection of SNHL during this
ity is a concern (eg, in those with impaired report monitoring viral load to assist in period is also most likely to improve long-
renal function) or where there are concerns decisions regarding adequate drug dosing term outcomes.55 Monitoring should con-
about treatment response. and detection of potential drug resistance; tinue into early childhood, however, because
however, most experts at this meeting deterioration in hearing continues throughout
Full Consensus Within This Expert do not conduct this routinely. Treatment early life55 (Quality B, Strength 1).
Group Was That duration is not altered by any viral param- Neurodevelopmental follow-up is
eters, and rebound of virus after treatment suggested at 1 and 2 years of age ideally with
1. Where treatment is given, babies should discontinuation is well documented with formal neurodevelopmental assessment. This
have regular weight measurement and no demonstrable association with long- is not, however, routinely conducted in all
safety monitoring to enable appropri- term outcomes (Quality D, Strength 2). If centers, and there is no evidence-based ben-
ate dose adjustment of medication (see viral load is checked after discontinuing efit in this particular group, although early
Table 5).(Quality A, Strength 1) drug, it is suggested that parents are fore- detection of functional impairments is gener-
2. Where treatment is given, parents should warned of the likelihood that virus will be ally agreed to be beneficial.
be fully counseled about both the known detectable and that this is of unknown sig- Ophthalmic follow-up is recom-
and potential side effects of treatment with nificance. mended annually at least until children can
current antivirals. (Quality A, Strength 1 talk in those with clinically detectable disease
for short-term side effects; Long-term, no at birth, but not in those without, because
published studies) FOLLOW-UP deterioration in vision has been observed in
3. Although there are theoretical risks of Table 5 summarizes recommended this group (Quality C, Strength 1).6
longer term treatment toxicity, no large follow-up of babies with cCMV (both treated Families should be given information
cohorts have been followed up to enable and untreated). for local/national support groups where these
this to be fully evaluated in humans treated The recommendation for audiologic exist (see acknowledgements). Where cCMV
during early life. Where possible, children follow-up is based on long-term surveillance parent groups are not easily accessible, par-
receiving antiviral treatment should, there- studies of SNHL in cCMV.4,16 Frequent fol- ents of children with hearing loss may find
fore, be entered into a registry to enable low-up is suggested during the first 2 years of support from groups for those with hearing
ongoing pharmacovigilance. life because this is the period of highest risk impairment.

© 2017 Wolters Kluwer Health, Inc. All rights reserved. www.pidj.com | 1211

Copyright © 2017 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
Luck et al The Pediatric Infectious Disease Journal  •  Volume 36, Number 12, December 2017

RECOMMENDATIONS FOR Recent publications have shown potential cost caused by congenital cytomegalovirus infection
be missed? J Pediatr. 1999;135:60–64.
FUTURE PEDIATRIC RESEARCH effectiveness of screening at birth for cCMV,
although these calculations are constrained by 5. Boppana SB, Pass RF, Britt WJ, et al. Symptomatic
congenital cytomegalovirus infection: neonatal
1. Clinical trials addressing treatment of the issues raised in this article regarding true morbidity and mortality. Pediatr Infect Dis J.
those with more “minor” manifestations quantification of benefits of treatment and 1992;11:93–99.
of disease/no clinically detectable disease agreed treatment duration in certain patient 6. Coats DK, Demmler GJ, Paysse EA, et al.
at birth and those with isolated SNHL. groups.56,57 It will be challenging to address Ophthalmologic findings in children with con-
2. Clinical studies of antenatal therapies to many of the research questions raised through genital cytomegalovirus infection. J AAPOS.
RCTs, given the significant resources and 2000;4:110–116.
decrease transmission of infection and
cCMV disease once infection is estab- long-term follow-up required alongside poten- 7. Kimberlin DW, Lin CY, Sánchez PJ, et al.;
tial difficulties in recruiting into such studies National Institute of Allergy and Infectious
lished. Diseases Collaborative Antiviral Study Group.
3. Publications relating to cCMV should when treatment is anecdotally being offered Effect of ganciclovir therapy on hearing in symp-
make it clear how those included were more freely. Collecting accurate data on dis- tomatic congenital cytomegalovirus disease
identified (ie, babies presenting with clini- ease manifestations and treatment outcomes involving the central nervous system: a rand-
in different patient groups alongside maternal omized, controlled trial. J Pediatr. 2003;143:16–
cally detected “symptoms” vs “screened” 25.
babies identified through existing antena- demographics can, however, inform treatment
8. Nassetta L, Kimberlin D, Whitley R. Treatment
tal or postnatal screening pathways includ- strategies as previously shown very effec- of congenital cytomegalovirus infection: impli-
ing hearing screening programs), or after tively for the management of pediatric human cations for future therapeutic strategies. J
further investigation of abnormalities, immunodeficiency virus. This requires a uni- Antimicrob Chemother. 2009;63:862–867.
such as thrombocytopenia, found inciden- fied approach to initial diagnostic tests, defini- 9. Dreher AM, Arora N, Fowler KB, et al. Spectrum
tally when blood sampling is performed tions of symptomatology and follow-up which of disease and outcome in children with symp-
is currently being addressed by a network of tomatic congenital cytomegalovirus infection. J
for other indications. Pediatr. 2014;164:855–859.
4. Development of clinical prediction mod- clinicians with an interest in this area through
10. Ancora G, Lanari M, Lazzarotto T, et al. Cranial
els to better categorize severity of disease both national and European initiatives such as ultrasound scanning and prediction of outcome in
(CNS vs non-CNS and babies with single Paediatric European Network for Treatment newborns with congenital cytomegalovirus infec-
vs multiple findings of disease) and asso- of AIDS - Infectious Diseases, the European tion. J Pediatr. 2007;150:157–161.
ciated outcomes to assist counseling of Congenital CMV Initiative and European 11. Amir J, Schwarz M, Levy I, et al. Is lenticulostri-
parents. Society of Paediatric Infectious Diseases ated vasculopathy a sign of central nervous sys-
and European Society for Clinical Virology tem insult in infants with congenital CMV infec-
5. Studies of neuroimaging, particularly tion? Arch Dis Child. 2011;96:846–850.
MRI, and added value with regards to pre- (ESCV). It should also be reiterated that this
12. de Jong EP, Lopriore E, Vossen AC, et al. Is rou-
dicting long-term impairments particu- article focuses on postnatal aspects of diagno- tine TORCH screening warranted in neonates
larly in those without clinically detectable sis and treatment. There is an associated and with lenticulostriate vasculopathy? Neonatology.
disease at birth through studies involving simultaneous need for work alongside obstet- 2010;97:274–278.
unselected cCMV cohorts. ric and fetal medicine colleagues to address 13. Alarcon A, Martinez-Biarge M, Cabañas F, et al.
6. Clinical trials of alternative treatment similar uncertainties in aspects of antenatal Clinical, biochemical, and neuroimaging find-
durations and new anti-CMV therapies care. It is hoped that through such collabo- ings predict long-term neurodevelopmental out-
come in symptomatic congenital cytomegalovirus
when available. rations, progress will be made in decreasing infection. J Pediatr. 2013;163:828–34.e1.
7. Biomarkers. It seems unlikely that a pre- infection and disease in fetuses, newborns and
14. Noyola DE, Demmler GJ, Nelson CT, et al.;

defined duration of treatment will be subsequently older children with cCMV. Houston Congenital CMV Longitudinal Study
similarly beneficial in babies with such Group. Early predictors of neurodevelopmental
varying clinical manifestations of dis- outcome in symptomatic congenital cytomegalo-
ACKNOWLEDGMENTS virus infection. J Pediatr. 2001;138:325–331.
ease and likely variable viral burden and The CMV-infected children and their
host immune function. The development 15. Gabrielli L, Bonasoni MP, Santini D, et al.

parents/carers who we have cared for and Human fetal inner ear involvement in congeni-
of both host and virologic biomarkers who challenge us on a daily basis to provide tal cytomegalovirus infection. Acta Neuropathol
of long-term outcomes would greatly evidence-based and consistent care including Commun. 2013;1:63.
enhance design of future RCTs and enable those parents who are involved with support- 16. Goderis J, De Leenheer E, Smets K, et al. Hearing
more accurate counseling and resource ing other parents through local or national loss and congenital CMV infection: a systematic
allocation. review. Pediatrics. 2014;134:972–982.
support groups including (but not exclu-
8. All children receiving treatment should sively) CMV Action (www.cmvaction.org.uk); 17. Jones CA. Congenital cytomegalovirus infec-

be captured in a registry to enable ongo- tion. Curr Probl Pediatr Adolesc Health Care.
https://www.stopcitomegalovirus.org. 2003;33:70–93.
ing pharmacovigilance for any long-term
effects of antiviral medication. 18. Benoist G, Leruez-Ville M, Magny JF, et al.

REFERENCES Management of pregnancies with confirmed cyto-
9. Identification of risk factors for maternal 1. Dollard SC, Grosse SD, Ross DS. New estimates megalovirus fetal infection. Fetal Diagn Ther.
virus transmission, particularly, in those of the prevalence of neurological and sensory 2013;33:203–214.
mothers with previous known exposure to sequelae and mortality associated with congeni- 19. Khan NA, Kazzi SN. Yield and costs of screening
CMV (CMV IgG seropositive). tal cytomegalovirus infection. Rev Med Virol. growth-retarded infants for torch infections. Am J
2007;17:355–363. Perinatol. 2000;17:131–135.
2. Morton CC, Nance WE. Newborn hearing 20. Lorenzoni F, Lunardi S, Liumbruno A, et al.

CONCLUSIONS screening–a silent revolution. N Engl J Med. Neonatal screening for congenital cytomegalo-
2006;354:2151–2164. virus infection in preterm and small for gesta-
As stated at the outset, this article repre-
sents the consensus opinion of a group of pro- 3. Guyatt GH, Oxman AD, Vist GE, et al.; GRADE tional age infants. J Matern Fetal Neonatal Med.
Working Group. GRADE: an emerging consen- 2014;27:1589–1593.
fessionals with a particular interest in cCMV. sus on rating quality of evidence and strength of 21. van der Weiden S, de Jong EP, Te Pas AB, et al. Is
It highlights that much of our practice is based recommendations. BMJ. 2008;336:924–926. routine TORCH screening and urine CMV culture
on limited data but identifies areas where there 4. Fowler KB, Dahle AJ, Boppana SB, et al. Newborn warranted in small for gestational age neonates?
is nonetheless consensus amongst experts. hearing screening: will children with hearing loss Early Hum Dev. 2011;87:103–107.

1212 | www.pidj.com © 2017 Wolters Kluwer Health, Inc. All rights reserved.

Copyright © 2017 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
The Pediatric Infectious Disease Journal  •  Volume 36, Number 12, December 2017 Congenital CMV: European Consensus

22. Vaudry W, Rosychuk RJ, Lee BE, et al. Congenital 34. Lanari M, Lazzarotto T, Venturi V, et al. Neonatal 45. Dahle AJ, Fowler KB, Wright JD, et al.

cytomegalovirus infection in high-risk Canadian cytomegalovirus blood load and risk of sequelae Longitudinal investigation of hearing disorders in
infants: report of a pilot screening study. Can J in symptomatic and asymptomatic congenitally children with congenital cytomegalovirus. J Am
Infect Dis Med Microbiol. 2010;21:e12–e19. infected newborns. Pediatrics. 2006;117:e76– Acad Audiol. 2000;11:283–290.
23. Turner KM, Lee HC, Boppana SB, et al. Incidence e83. 46. Rivera LB, Boppana SB, Fowler KB, et al.

and impact of CMV infection in very low birth 35. Ross SA, Novak Z, Fowler KB, et al.
Predictors of hearing loss in children with symp-
weight infants. Pediatrics. 2014;133:e609–e615. Cytomegalovirus blood viral load and hearing tomatic congenital cytomegalovirus infection.
24. Luck S, Sharland M. Postnatal cytomegalovirus: loss in young children with congenital infection. Pediatrics. 2002;110:762–767.
innocent bystander or hidden problem? Arch Dis Pediatr Infect Dis J. 2009;28:588–592. 47. Bilavsky E, Shahar-Nissan K, Pardo J, et al.

Child Fetal Neonatal Ed 2009; 94(1):F58–F64. 36. Luck SE, Emery VC, Atkinson C, et al.
Hearing outcome of infants with congenital cyto-
Compartmentalized dynamics of cytomegalovi- megalovirus and hearing impairment. Arch Dis
25. de Vries JJ, van Zwet EW, Dekker FW, et al. The
rus replication in treated congenital infection. J Child. 2016;101:433–438.
apparent paradox of maternal seropositivity as a
risk factor for congenital cytomegalovirus infec- Clin Virol. 2016;82:152–158. 48. Amir J, Attias J, Pardo J. Treatment of late-onset
tion: a population-based prediction model. Rev 37. Goycochea-Valdivia WA, Baquero-Artigao F,
hearing loss in infants with congenital cyto-
Med Virol. 2013;23:241–249. Del Rosal T, et al.; Spanish Registry of Infants megalovirus infection. Clin Pediatr (Phila).
with Congenital Cytomegalovirus Infection 2014;53:444–448.
26. Wang C, Zhang X, Bialek S, et al. Attribution of
congenital cytomegalovirus infection to primary (REDICCMV) Study Group. Cytomegalovirus 49. del Rosal T, Baquero-Artigao F, Blázquez D, et al.
versus non-primary maternal infection. Clin DNA detection by polymerase chain reaction Treatment of symptomatic congenital cytomeg-
Infect Dis. 2011;52:e11–e13. in cerebrospinal fluid of infants with congenital alovirus infection beyond the neonatal period. J
infection: associations with clinical evaluation at Clin Virol. 2012;55:72–74.
27. de Vries JJ, van der Eijk AA, Wolthers KC, et al. birth and implications for follow-up. Clin Infect
Real-time PCR versus viral culture on urine as a 50. Kimberlin DW, Acosta EP, Sánchez PJ, et al.;
Dis. 2017;64:1335–1342. National Institute of Allergy and Infectious
gold standard in the diagnosis of congenital cyto-
megalovirus infection. J Clin Virol. 2012;53:167– 38. Capretti MG, Lanari M, Tani G, et al. Role of cer- Diseases Collaborative Antiviral Study Group.
170. ebral ultrasound and magnetic resonance imaging Pharmacokinetic and pharmacodynamic assess-
in newborns with congenital cytomegalovirus ment of oral valganciclovir in the treatment of
28. Boppana SB, Ross SA, Shimamura M, et
infection. Brain Dev. 2014;36:203–211. symptomatic congenital cytomegalovirus disease.
al.; National Institute on Deafness and Other J Infect Dis. 2008;197:836–845.
Communication Disorders CHIMES Study. 39. de Vries LS, Gunardi H, Barth PG, et al. The
Saliva polymerase-chain-reaction assay for cyto- spectrum of cranial ultrasound and magnetic 51. Gwee A, Curtis N, Connell TG, et al. Ganciclovir
megalovirus screening in newborns. N Engl J resonance imaging abnormalities in congeni- for the treatment of congenital cytomegalovirus:
Med. 2011;364:2111–2118. tal cytomegalovirus infection. Neuropediatrics. what are the side effects? Pediatr Infect Dis J.
2004;35:113–119. 2014;33:115.
29. Koyano S, Inoue N, Nagamori T, Moriuchi H,
Azuma H. Newborn screening of congenital cyto- 40. Manara R, Balao L, Baracchini C, et al. Brain 52. Valcyte Safety Data. Available at https://www.

megalovirus infection using saliva can be influ- magnetic resonance findings in symptomatic gene.com/download/pdf/valcyte_prescribing.pdf.
enced by breast feeding. Arch Dis Child Fetal congenital cytomegalovirus infection. Pediatr Accessed 13 March 2017.
Neonatal Ed 2013; 98(2):F182. Radiol. 2011;41:962–970. 53. Tomicic MT, Bey E, Wutzler P, et al. Comparative
30. Ross SA, Ahmed A, Palmer AL, et al.; National 41. Kimberlin DW, Jester PM, Sánchez PJ, et al.;
analysis of DNA breakage, chromosomal aber-
Institute on Deafness and Other Communication National Institute of Allergy and Infectious rations and apoptosis induced by the anti-herpes
Disorders CHIMES Study. Detection of congeni- Diseases Collaborative Antiviral Study Group. purine nucleoside analogues aciclovir, ganciclovir
tal cytomegalovirus infection by real-time poly- Valganciclovir for symptomatic congeni- and penciclovir. Mutat Res. 2002;505:1–11.
merase chain reaction analysis of saliva or urine tal cytomegalovirus disease. N Engl J Med. 54. Scott JC, Partovi N, Ensom MH. Ganciclovir in
specimens. J Infect Dis. 2014;210:1415–1418. 2015;372:933–943. solid organ transplant recipients: is there a role
31. Yamamoto AY, Mussi-Pinhata MM, Marin LJ,
42. Luck S, Sharland M, Griffiths P, et al. Advances for clinical pharmacokinetic monitoring? Ther
et al. Is saliva as reliable as urine for detection in the antiviral therapy of herpes virus infec- Drug Monit. 2004;26:68–77.
of cytomegalovirus DNA for neonatal screen- tion in children. Expert Rev Anti Infect Ther. 55. Davis A, Bamford J, Wilson I, et al. A critical
ing of congenital CMV infection? J Clin Virol. 2006;4:1005–1020. review of the role of neonatal hearing screening in
2006;36:228–230. 43. Mareri A, Lasorella S, Iapadre G, et al. Anti-viral the detection of congenital hearing impairment.
32. Wang L, Xu X, Zhang H, et al. Dried blood spots therapy for congenital cytomegalovirus infec- Health Technol Assess. 1997;1:i–iv, 1.
PCR assays to screen congenital cytomegalovirus tion: pharmacokinetics, efficacy and side effects. 56. Gantt S, Dionne F, Kozak FK, et al. Cost-

infection: a meta-analysis. Virol J. 2015;12:60. J Matern Fetal Neonatal Med. 2016;29:1657– effectiveness of universal and targeted newborn
33. Bradford RD, Cloud G, Lakeman AD, et al.;
1664. screening for congenital cytomegalovirus infec-
National Institute of Allergy and Infectious
44. Oliver SE, Cloud GA, Sánchez PJ, et al.; tion. JAMA Pediatr. 2016;170:1173–1180.
Diseases Collaborative Antiviral Study Group. National Institute of Allergy, Infectious 57. Williams EJ, Gray J, Luck S, Atkinson C,

Detection of cytomegalovirus (CMV) DNA by Diseases Collaborative Antiviral Study Group. Embleton ND, Kadambari S et al. First estimates
polymerase chain reaction is associated with Neurodevelopmental outcomes following ganci- of the potential cost and cost saving of protect-
hearing loss in newborns with symptomatic con- clovir therapy in symptomatic congenital cytomeg- ing childhood hearing from damage caused by
genital CMV infection involving the central nerv- alovirus infections involving the central nervous congenital CMV infection. Arch Dis Child Fetal
ous system. J Infect Dis. 2005;191:227–233. system. J Clin Virol. 2009;46(suppl 4):S22–S26. Neonatal Ed 2015; 100(6):F501–F506.

© 2017 Wolters Kluwer Health, Inc. All rights reserved. www.pidj.com | 1213

Copyright © 2017 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.