Best Practice & Research Clinical Obstetrics and Gynaecology xxx (2016) 1e11

Contents lists available at ScienceDirect

Best Practice & Research Clinical

Obstetrics and Gynaecology
journal homepage: www.elsevier.com/locate/bpobgyn

11

Fetal cytomegalovirus infection

Marianne Leruez-Ville, M.D., PhD. a, b,
Yves Ville, M.D., FRCOG a, c, *
a
EA 73-28, Universit

e Paris Descartes, Sorbonne Paris Cit

e, Paris, 75005, France
b ^pital Necker-E.M., Laboratoire de Microbiologie Clinique. Centre national de R

AP-HP, Ho ef 
erence
Cytomegalovirus- Laboratoire associ
e, Paris, 75015, France
c ^pital Necker-E.M., Maternit

AP-HP, Ho e, Unit
edecine Fœtale, Paris, 75015, France
e de M

Cytomegalovirus (CMV) congenital infection affects 0.7% of live

Keywords:
births worldwide and is the leading cause of congenital neuro-
cytomegalovirus
fetus logical handicap of infectious origin. However, systematic
serology screening for this infection has not been implemented in preg-
imaging nancy or at birth in any country. This apparent paradox had been
hyperimmune globulin justified by persisting gaps in the knowledge of this congenital
valaciclovir infection: uncertain epidemiological data, difficulty in the diag-
nosis of maternal infection, absence of validated prenatal prog-
nostic markers, unavailability of an efficient vaccine and scarcity of
data available on the treatment. However, in the last decade, new
data have emerged towards better management of this congenital
infection, including solid epidemiological data, good evidence for
the accuracy of diagnosis of maternal CMV infection and good
evidence for the feasibility of predicting the outcome of fetal
infection by a combination of fetal imaging and fetal laboratory
parameters. There is also some evidence that valaciclovir treat-
ment of mothers carrying an infected fetus is feasible, safe and
might be effective. This review provides an update on the evidence
for diagnosis, prognosis and treatment of congenital infection in
the antenatal period. These suggest a benefit to a proactive
approach for prenatal congenital infections.
© 2016 Published by Elsevier Ltd.

* Corresponding author. Ho ^ pital Necker-E.M., Unite

de Me

decine Fœtale, 149 rue de Se
vres, 75015 Paris, France.
Tel.: þ33 1 44 49 49 62; Fax: þ33 1 44 49 49 60.
E-mail address: ville.yves@gmail.com (Y. Ville).

http://dx.doi.org/10.1016/j.bpobgyn.2016.10.005
1521-6934/© 2016 Published by Elsevier Ltd.

Please cite this article in press as: Leruez-Ville M, Ville Y, Fetal cytomegalovirus infection, Best Practice &
Research Clinical Obstetrics and Gynaecology (2016), http://dx.doi.org/10.1016/j.bpobgyn.2016.10.005
2 M. Leruez-Ville, Y. Ville / Best Practice & Research Clinical Obstetrics and Gynaecology xxx (2016) 1e11

Epidemiology of CMV fetal infection

The epidemiology of congenital cytomegalovirus infection has been described by numerous

concordant studies published over the last thirty years and is now well established (level III).
Congenital CMV (cCMV) infection has a birth prevalence of 0.7% worldwide [1]. However, birth
prevalence varies according to the level of CMV immunity in pregnant women. The birth prevalence is
the highest (>1%) in countries where maternal seroprevalence is high (>95%) such as most African and
Asian countries. In countries with low maternal seroprevalence (around 50%) such as most European
countries, the birth prevalence is the lowest around 0.4%. This reflects the burden of cCMV following
maternal secondary infection. Indeed, fetal infection may follow maternal primary infection during
pregnancy but also maternal secondary infection which happens in women already immune for CMV
before pregnancy. The ratio between fetal infections related to maternal primary infections and those
related to maternal secondary infections depends on the seroprevalence rate in pregnant women. In
highly seroimmune populations, almost all fetal infections are related to secondary maternal infections
[2]. In countries with intermediate CMV seroprevalence (z70%) such as the US, 25% of congenital
infections are expected to be related to primary maternal infections and the other 75% to maternal
secondary infections [3]. In countries with low maternal seroprevalence such as most European
countries (z50%) an equal ratio between fetal infections related to primary infection and those related
to maternal secondary infection is expected [4,5].
In a meta-analysis based on 15 studies with a total of 117,986 infants screened, the percentage of
infected children with CMV-specific symptoms at birth was 12.7%, the percentage of symptomatic
children with permanent sequelae was 40-58% and the percentage of children without symptoms at
birth who developed permanent sequelae was estimated to be 13.5% [1]. In most studies the definition
of a symptomatic neonate is a neonate with clinical and/or laboratory abnormalities (petechia,
microcephaly, hepatosplenomegaly, fetal growth restriction (FGR), chorioretinitis, thrombocytopenia,
hepatitis and hearing loss). In older studies, hearing loss was not tested at birth and this criterion was
not included in the definition of a symptomatic neonate. A first study reported that severe symptoms at
birth and long-term sequelae were seen much more frequently among neonates infected after a
maternal primary infection than in those infected after a secondary infection [6]. However, the results
of early first report have been challenged by more recent data from longitudinal studies published by
the same group and by others, indicating that the fetal morbidity related to maternal secondary
infection is in fact as high as that related to maternal primary infection. In a series of 300 neonates (176
infected after maternal primary infection and 124 after maternal secondary infection), the proportion
of symptomatic neonates was 11% in both groups and the proportion of hearing loss was also 10% in
both groups [7]. In another long-term follow-up study, non-primary infections contributed substan-
tially to the burden of childhood congenital CMV disease: half of the children presenting with mod-
erate to severe outcomes were born from mothers with non-primary infections [8].
The epidemiology and pathophysiology of maternal secondary infections are not well under-
stood.(level I) They could be related to either a reactivation of the endogen CMV strain or to a rein-
fection with a new CMV strain and the relative contribution of both mechanisms to the burden of
secondary infection is unknown. The high genetic variability of the CMV genome with 4.7% variability
between strains at nucleotide level could explain the possibility of re-infection [9]. The likelihood of re-
infection in pregnant women was suggested by the evidence of the acquisition of new CMV antibody
specificities between serum samples collected before and after pregnancy in seropositive women who
delivered an infected neonate [10]. (level II).

Diagnosis of maternal infection

The diagnosis of primary CMV infection in pregnancy is ascertained when seroconversion is

documented with the presence of CMV specific IgG in the serum of a pregnant women who was
previously tested IgG negative in a serum, sample collected earlier in pregnancy. (level III) However, the
opportunity to diagnose seroconversion is rarely feasible in the absence of screening and prospective
monitoring.

Please cite this article in press as: Leruez-Ville M, Ville Y, Fetal cytomegalovirus infection, Best Practice &
Research Clinical Obstetrics and Gynaecology (2016), http://dx.doi.org/10.1016/j.bpobgyn.2016.10.005
 M. Leruez-Ville, Y. Ville / Best Practice & Research Clinical Obstetrics and Gynaecology xxx (2016) 1e11 3

In the absence of seroconversion and based on clinical experience, the recommended best practice
for the diagnosis of CMV primary infection is to perform a serology test with measurement of both
specific IgM and IgG [11]. In very rare cases, IgM response may be transient with CMV-IgM antibodies
persisting for a very short time and consequently undetected even in the context of a recent primary
infection. However, in most primary infections IgM antibodies reach moderate to high levels and
persist for weeks, so are unlikely to be missed by standard commercial kits. Therefore, the sensitivity of
screening recent primary infections by IgM testing is very high. The main problem is to deal with the
low specificity of IgM screening. There are numerous situations in which detection of CMV IgM is not
related to a recent primary infection such as 1) persistent specific IgM that can be detect for months in
some individuals, 2) IgM cross relativities related to other ongoing viral infection or to non-specific
stimulation of the immune system, 3) rise in IgM levels in episodes of reactivation. The interpreta-
tion of a positive IgM test must therefore be cautious and the state of the art is to perform IgG avidity to
discriminate between primary infection during or prior to the current pregnancy [12e15]. CMV IgG
avidity testing is now a proven, valuable laboratory tool for diagnosing primary CMV infection during
pregnancy. This assay is based on the observation that CMV specific low avidity IgG are produced
during the first months of infection whereas binding avidity of maternal IgG to CMV antigens increases
over time and high avidity IgG are detected in remote infection. The avidity index expresses the per-
centage of IgG still bound to the CMV antigens following treatment with denaturing agents. In com-
mercial avidity assays, avidity indices are classified in 3 categories according to their values: low avidity
which is in favour of a recent primary infection within the last 3 months, high avidity which excludes
the onset of a primary infection in the last 3 months or intermediate avidity which does not allow
discrimination between recent or remote primary infection. Earlier studies reported a low concordance
between avidity commercial assays [16,17] whereas new generations of CMV avidity assays exhibit a
better standardisation [18,19]. In different studies evaluating transmission rates in relation to avidity
results, around 30% of women with a low avidity value transmitted CMV to their offspring; in contrast,
none of the women with high avidity during the first trimester transmitted CMV infection, and women
with an intermediate-avidity result during the first trimester had a low risk of intrauterine trans-
mission of around 5% [17,20e22]. These findings have significant implications for prenatal counselling
1) women with high CMV IgG avidity during the first trimester can be assured that the risk of giving
birth to an infected infant is low and invasive procedures to identify fetal infection are not needed, 2)
whereas women with low avidity should be considered for further testing to assess fetal infection
status. (level II).
One study reported that the presence of CMV DNA in maternal blood significantly increases the risk
of viral transmission to the fetus in women with a primary infection [17], and viremia combined with
avidity value allows calculation of the incremental risk of fetal transmission. Studies are needed to
confirm these findings and to design an accurate model based on a combination of laboratory tests (IgG
levels, IgM levels, IgG avidity and CMV PCR in maternal blood and urine) to predict which pregnancies
would result in an infected fetus.
There are no validated laboratory serological or virological tools to diagnose maternal secondary
infections.

Diagnosis of fetal infection

Both CMV primary infections in pregnancy and the presence of compatible ultrasound fetal signs
are opportunities for diagnosis of CMV infection in the fetus.
From a series of studies published in the early 2000's, detection of viral DNA by polymerase chain
reaction (PCR) in the amniotic fluid has been recognized as the reference method for the diagnosis of
fetal infection [23e26]. (level III) Although virus isolation by culture in amniotic fluid has an absolute
specificity it has a lower sensitivity than PCR and is no longer the reference method. Earlier studies
based on traditional PCR method such as nested-PCR reported some false positive PCR results, chal-
lenging the specificity of CMV PCR in amniotic fluid. These false positive results are probably explained
by laboratory contaminations which happened in those early days. Today the generalisation of auto-
mated real time PCR methods has helped to overcome the risk of contamination and achieve quasi-
absolute specificity for prenatal diagnosis of CMV infection by PCR. The sensitivity of CMV PCR in

Please cite this article in press as: Leruez-Ville M, Ville Y, Fetal cytomegalovirus infection, Best Practice &
Research Clinical Obstetrics and Gynaecology (2016), http://dx.doi.org/10.1016/j.bpobgyn.2016.10.005
4 M. Leruez-Ville, Y. Ville / Best Practice & Research Clinical Obstetrics and Gynaecology xxx (2016) 1e11

amniotic fluid is also excellent when timing of amniocentesis is appropriate. Following maternal pri-
mary infection, the process leading to CMV excretion in the fetal urine will take an average of 6-8
weeks and this interval should be recognised in order to avoid false negative prenatal diagnosis [20,23].
Amniocentesis should also be performed once fetal urination is well established and therefore not
before 20 weeks.(level III) When the conditions of sampling are optimal, the sensitivity of prenatal
diagnosis by PCR in amniotic fluid has been reported to be between 90 to 95%. These 5% to 10% of false
negative prenatal diagnosis cases (neonates born infected following after a negative prenatal diag-
nosis) are explained by a late transplacental passage of the virus, later than 6 to 8 weeks after primary
maternal infection. In a recent retrospective case control study, long-term outcomes at 2 years of age
were compared between infants born after false amniocentesis (study group) and those with a positive
amniocentesis (control group) [27]. There were 0% (0/46) sequelae in the study group compared to 14%
(13/91) in the control group. The reason for the good outcome in those cases with a false negative
prenatal diagnosis could be: 1) the later transmission of the virus recognising that fetal infection at a
later stage of pregnancy is likely to carry a better prognosis than fetal infection in the first trimester
([28,29], 2) protection by an already mature maternal immunity that could be incapable of preventing
fetal infection but capable of preventing fetal disease.
There have been concerns over the potential iatrogenic risk of CMV transmission during antenatal
invasive procedures in the presence of viral DNA in maternal blood. However, two well-designed
studies with significant sample size demonstrated that this risk was negligible. In these studies, the
proportions of infected neonates were similar in mothers with negative CMV DNA detection in blood at
the time of amniocentesis and in those with positive CMV DNA detection in blood at that time [11].
Therefore, amniocentesis can safely be done in cases with concomitant positive DNA detection in the
maternal blood. (level II).

Prognostic markers of fetal infection

Once the diagnosis of fetal infection is proven, the prognosis is established based on a combination
of fetal imaging and fetal laboratory tests.

Imaging findings

(Table 1) The pathophysiology of fetal CMV infection leads one to expect progressive and sometimes
only subtle or transient findings on ultrasound. CMV is a viraemic herpes virus that will reach the fetus
via the umbical circulation. The placenta, where the virus replicates, will act as a barrier against CMV
but also as a reservoir. Within 4 to 8 weeks following maternal viraemia, an early but inconstant sign of
vertical transmission is therefore likely to be placentitis as defined by a thickness of 4 cm or more and a
heterogeneous appearance typically with calcifications co-existing with hypoechoic areas.
Once the virus reaches the fetal circulation, the fetal kidneys are affected early and preferentially
which may cause transient oligohydramnios and less often renal hyperechogenicity. This appears to be
more frequent than polyhydramnios in fetal CMV infection [30]. Viral enterocolitis is often apparent
with the transient or persistent appearance of at least grade-2 hyper-echogenic bowel as a common
and early ultrasound finding [31]. This usually represents meconium ileus or bowel perforation with
meconial peritonitis even in the absence of ascites [32].
Several weeks can then elapse until other features of fetal infection, if any, show-up on prenatal
ultrasound, and some earlier features may have disappeared by then. Overt systemic disease might
show as hepato-splenomegaly and possibly ascites in the fetus as a result of cholestatic hepatitis and
liver insufficiency [33]. Less often, generalised oedema will suggest anaemia-related hydrops due to the
combined effect of liver failure and bone marrow infection. Cardiomyopathy, expressed as car-
diomegaly with a thick myocardium which may contain punctuate calcifications, is a rare finding which
could also contribute to the development of fetal hydrops and could be associated with tachy-
arrythmia [33]. Calcifications of the fetal liver, spleen, and even lungs may show and remain as a
marker of a systemic fetal disease [34].
Intrauterine fetal growth restriction (FGR) may develop as a result of either fetal infection or
placental infection or both. It can therefore be advisable to screen for CMV as part of the assessment of

Please cite this article in press as: Leruez-Ville M, Ville Y, Fetal cytomegalovirus infection, Best Practice &
Research Clinical Obstetrics and Gynaecology (2016), http://dx.doi.org/10.1016/j.bpobgyn.2016.10.005
 M. Leruez-Ville, Y. Ville / Best Practice & Research Clinical Obstetrics and Gynaecology xxx (2016) 1e11 5

Table 1
Ultrasound symptoms related to CMV.

Severe US brain abnormalities Mild US brain abnormalities Extra-cerebral US abnormalities

Ventriculomegaly* 15 mm Mild ventriculomegaly*(>10 to 15 mm) ****Hyperechogenic bowel

Periventricular hyperechogenicity Intra-ventricular adhesions ***Hepatomegaly (right lobe  40 mm)
Hydrocephalus** Intracerebral calcifications Splenomegaly (longest diameter  40 mm
in the second trimester)
Microcephaly < -2DS Subependymal cysts Intra-uterine growth retardation
(<5th centile)
Increased cisterna magna  8 mm Choroid plexus cysts Oligoamnios (deepest vertical
pool < 2.5 cm)
Vermian hypoplasia Calcifications of the lenticulostriate Polyhydramnios (deepest vertical
vessels in the basal ganglia pool > 10 cm)
Porencephaly Ascites
Lissencephaly Pleural effusion
Periventricular cystic lesions of Fetal hydrops, subcutaneous oedema
the white matter
Agenesis of the corpus callosum Placentomegaly  40 mm
Intra-hepatitic calcifications
*
Ventriculomegaly refers to increased measurement of the lateral ventricles at the level of glomus.
**
Hydrocephalus refers to tri- or quadri-ventricular dilatation in relation with micro-encephaly in this case.
***
The right lobe of the liver must be measured in a para-sagittal plane:aA measurement >40 mm in the second trimester is
considered abnormal.
****
Hyperechogenic bowel was only considered when the echogenicity of the bowel was equal or more intense than that of the
fetal bones.

any FGR fetus below the 5th centile. Indeed, this could be a completely isolated finding irrespective of
placental or fetal Doppler examination.
Damage to the fetal brain is a late finding that can show variable and progressive features on
prenatal imaging. These can be present while the non-cerebral features have resolved, weeks or
months after the onset of maternal and even of fetal infection. Microcephaly is a major form of the
disease; however, this may be a very difficult diagnosis to establish, especially in a growth restricted
fetus [35]. Ventriculomegaly, unilateral or bilateral, is a common clue to the diagnosis since around 5%
of all ventriculomegaly diagnosed in utero are of infectious origin [36]. It can be of 2 types, namely
destructive or obstructive. Destructive ventriculomegaly is commonly moderate and will often precede
microcephaly, showing even subtle enlargement of pericerebral spaces as an early sign of micro-
encephaly. Obstructive ventriculomegaly can occur as a result of obstruction of the foramen of Mon-
roe and/or of Magendie and Lushka as a consequence of ventriculitis-related oedema or of intra-
ventricular haemorrhage [37]. Extreme forms can mimick acqueductal stenosis-related hydrocepha-
lus. The same mechanisms can lead to less common presentations such as mega-cisterna magna,
cerebellar hypoplasia or haemorrhage, pseudo-Dandy Walker malformations and schizencephaly [38].
More subtle anomalies can be identified as associated findings of any of the features described
above or as isolated findings, making the diagnosis more difficult. Non-specific vasculitis in the fetal
thalami and basal ganglia [39], described as candle-stick images, punctuate echogenicity within the
brain parenchyma and underlying the rim of the lateral ventricles together with strands within the
lateral ventricles. Germinolysis-related sub-ependymal cysts can also be overlooked by routine fetal
ultrasound examination when fetal infection is unknown [38]. Rare cases of clastic abnormalities of the
corpus callosum have been described in utero and at birth [38].
Abnormal myelination and gyration of the fetal brain is another pitfall for fetal brain ultrasound
examination and the development of fetal MRI is a definite asset in the complete assessment of high-
risk fetuses [40e42]. Lissencephaly could reflect injury before 16 or 18 weeks' whereas polymicrogyria
could reflect injury at 18 to 24 weeks. Cases with normal gyral patterns are probably injured during the
third trimester and show diffuse heterogenity in the white matter. Both T1 and T2 sequences are
therefore useful.

Please cite this article in press as: Leruez-Ville M, Ville Y, Fetal cytomegalovirus infection, Best Practice &
Research Clinical Obstetrics and Gynaecology (2016), http://dx.doi.org/10.1016/j.bpobgyn.2016.10.005
6 M. Leruez-Ville, Y. Ville / Best Practice & Research Clinical Obstetrics and Gynaecology xxx (2016) 1e11

Prognostic value of imaging findings

Severe cerebral lesions have been consistently associated in the literature with a dismal prognosis
[43], with the odds ratio for a poor outcome as high as 40.6 [44]. Therefore termination of pregnancy
should be discussed in those cases. In several well documented series, a normal ultrasound exami-
nation was associated with a residual risk of severe infection at birth including deafness and more
severe neurodevelopmental abnormalities, of 1-5% and 0-5%, respectively [41,44e47]. These risk es-
timates were based upon imaging, mainly ultrasound alone, throughout the pregnancy. Indeed, the
brain involvement might be delayed up until late in pregnancy and the prognostic value of ultrasound
at diagnosis is expected to be lower than the prognostic value of imaging obtained later in pregnancy.
However, in a recent study the negative predictive value of ultrasound features noted at the time of
diagnosis was as high as 93% [48]. The predictive value of non-severe ultrasound symptoms has been
evaluated in only 2 studies, with odds ratios for a poor outcome when showing extra-cerebral
symptoms of between 7.4 and 18.4, with a positive predictive value of 54% to 60% [44,48].
Although target ultrasound has a good negative predictive value when the operator is aware of fetal
infection, it should be stressed that systematic ultrasound performed as part of routine antenatal care
has a poor sensitivity, as low as 35% [45]. Therefore, in cases with no ultrasound signs and in those with
non-severe ultrasound signs, other prognostic markers are needed to refine the prognosis. The addition
of fetal cerebral MRI to serial targeted ultrasound examination in cases of proven fetal infection allows
negative predictive values of around 90% [40,41].

Prognostic value of fetal laboratory parameters

Viral load in amniotic fluid

The prognosis value of CMV DNA loads in the amniotic fluid is controversial. Most studies failed to
demonstrate an association between high CMV DNA load in amniotic fluid and greater risk of symp-
tomatic infection in the fetus or in the newborn [49e52]. In these studies, the results were not adjusted
for gestational age at amniocentesis or for the presumed date of maternal infection, probably
explaining this lack of association. Indeed, there is a significant positive correlation between the level
of CMV DNA in amniotic fluid and both gestational age at sampling and the time interval between
maternal primary infection and amniotic fluid sample collection [50,51]. This increase in CMV DNA
loads in amniotic fluid over time during pregnancy and after primary maternal infection could be
explained by the accumulation of CMV DNA in amniotic fluid as well as by the enhanced urinary flow of
the fetus throughout pregnancy. In one study, CMV DNA load in amniotic fluid was adjusted to the
interval time between amniocentesis and maternal primary infection in order to overcome the effect of
DNA accumulation. This showed that higher adjusted CMV DNA loads in amniotic fluid were signifi-
cantly associated with a symptomatic status at birth. Therefore, in this study the presence of non-
severe ultrasound symptoms associated with a CMV DNA level in amniotic fluid > 1M0M had a 78%
positive predictive value and a 90% negative predictive value for a symptomatic status at birth [48].

Laboratory parameters in fetal blood

Fetal blood sampling by cordocentesis for diagnostic purposes, particularly karyotyping, has fallen
into disuse and has largely been replaced by amniocentesis, hence reducing its practice to a few centres
that are involved in fetal blood transfusion. However, blood analysis in cCMV infection has proven
useful for the assessment of infected neonates [53,54]. Similarly, its application to the fetus has shown
that both thrombocytopenia (less than 50,000/mm3 in one study and less than 100,000/mm3 in two
studies) and a high viral load in fetal blood (over  4.93 log10 IU/ml in one study and 4.5 log10 copies/ml
in the other) were associated with a greater risk of giving birth to a symptomatic neonate or leading to
termination of pregnancy for severe brain abnormalities [44,48,55]. The cut-off values reported here
for platelet count and for CMV DNA load may therefore be used in clinical decision-making algorithms.
However, it should be stressed that even if standardized units are used, the implementation of CMV
DNA load cut-off values should be validated in different clinical and laboratory settings. In one study a
high level of ß2 microglobulin in fetal blood (>11.5 mg/l) was also strongly associated with a poor
outcome [55].

Please cite this article in press as: Leruez-Ville M, Ville Y, Fetal cytomegalovirus infection, Best Practice &
Research Clinical Obstetrics and Gynaecology (2016), http://dx.doi.org/10.1016/j.bpobgyn.2016.10.005
 M. Leruez-Ville, Y. Ville / Best Practice & Research Clinical Obstetrics and Gynaecology xxx (2016) 1e11 7

Fetal laboratory parameters should optimally be combined with imaging to establish the prognosis
at the time of positive amniocentesis. The association of either a high fetal blood viral load or a low
platelet counts with non-severe ultrasound features had a positive predictive value of 79% compared to
60% for non-severe ultrasound features alone [48]. The absence of high fetal blood viral load or of a low
platelet count together with the absence of non-severe ultrasound features had a negative predictive
value of 100% compared to 93% for the absence of non-severe ultrasound features only.

Treatment of fetal infection

Prevention of mother to fetus transmission by passive immunization with hyperimmune globulin

The clinical efficacy of hyperimmune globulin to prevent cCMV infection remains highly debated
after the publication of 2 main studies. Back in 2005, Nigro et al. [56] published the results of a non-
randomized clinical trial using intravenous CMV hyperimmune globulin in pregnant women with
maternal primary CMV infection. They included 102 women with a recent primary infection and un-
known fetal status before 21 weeks' gestation; 37 received hyperimmune globulin (100 U per kg
intravenously monthly) and 65 did not. Six of the 37 women (16%) who received hyperimmune globulin
had infants with congenital CMV infection, as compared with 19 of 47 women (40%) who did not receive
hyperimmune globulin. The authors concluded that hyperimmune globulin therapy was associated with
a significantly lower risk of congenital CMV infection (p<0.001) after maternal primary infection.
However, the results of this early study were not confirmed by a double-blind randomized, placebo-
controlled trial recently published [57]. In this phase 2 trial, hyperimmune globulin and placebo were
compared in 124 pregnancies with primary CMV infection at a median of 13 weeks and recruited
following systematic screening in the first trimester of pregnancy. Pregnant women were randomly
assigned within 6 weeks after the presumed onset of primary infection to receive either hyperimmune
globulin intravenously at the dose of 100U per kg of body weight or 0.9% saline solution every 4 weeks
up until 36 weeks or amniocentesis. Transmission rates were not significantly different between the two
groups, with 18 out of 61 (30%) and 27 out of 62 (44%) in the hyperimmune globulin and placebo groups,
respectively (p ¼ 0.13). In addition, there were 7 premature babies (15%) in the hyperimmune globulin
group and 1 in the placebo group (2%) (p ¼ 0.06). This trial that had been sized on the basis of the
benefits reported in the observational study by Nigro et al. [36] might have been underpowered. One
other randomized, placebo-controlled trial is currently being conducted in the United States
(NCT01376778 in clinicaltrials.gov). This study might hopefully be more conclusive on both the efficacy
and safety of hyperimmune globulin administration to prevent congenital CMV infection.
In the meantime, hyperimmune globulins treatment for pregnant woman with primary CMV
infection is not recommended. (level II).

Antiviral therapy for infected fetuses

There is only one study reporting the use of antiviral therapy in infected fetuses. In this study
valaciclovir was used to treat fetuses with non-severe ultrasound symptoms.
The rationale for using valaciclovir is three-fold:

1. Although aciclovir is not the best drug against CMV in vitro, with much higher 50% inhibiting
concentration than that of ganciclovir [58], it has been shown to be effective against CMV infection
in clinical settings. In renal transplantation and in HIV infection, high-dose regimens of valaciclovir
were effective in preventing CMV disease and suppressing CMV viraemia.
2. Acyclovir has the best safety profile among anti-CMV drugs: it is not genotoxic in vitro and no drug-
related increase of neoplasia has been observed in animals [59]. This good safety profile is reflected
by the reassuring safety data reported in pregnancy: a) in the pregnancy registry managed by the
manufacturer, the rate of birth defects in 596 pregnancies exposed to aciclovir in the first trimester
was 3.2%, thus in the same range as the expected rate in untreated pregnancies [60], b) and in a larger
study, based on a nationwide registry in Denmark, the rate of birth defects was not significantly
different between 1561 women exposed to aciclovir in the first trimester, 2379 women exposed to

Please cite this article in press as: Leruez-Ville M, Ville Y, Fetal cytomegalovirus infection, Best Practice &
Research Clinical Obstetrics and Gynaecology (2016), http://dx.doi.org/10.1016/j.bpobgyn.2016.10.005
8 M. Leruez-Ville, Y. Ville / Best Practice & Research Clinical Obstetrics and Gynaecology xxx (2016) 1e11

aciclovir in the second and third trimesters and 835,991 unexposed pregnant women (with rates of
birth defects at 2.0%, 2.0% and 2.4% respectively) [61]. Although neither the dose regimens of aciclovir
nor the length of the exposure were detailed in the Danish paper, these data are reassuring.
3. Valaciclovir can cross the second trimester placenta and a regimen of 8g/day given to the pregnant
woman orally achieved expected concentrations in maternal and fetal blood [62].

The valaciclovir trial was a phase II open label trial with only one arm based on a Simon optimal
two-stage design [63]. In this study, only mothers carrying an infected fetus presenting with non-
severe ultrasound features (extra-cerebral ultrasound abnormalities and/or mild ultrasound brain
abnormalities, (see Table 1), were included and treated with a regimen of 8g/day valaciclovir from
prenatal diagnosis until delivery or termination of pregnancy. The primary end point of this study was
the proportion of asymptomatic neonates born to treated mothers. The sample size (43 fetuses) was
calculated on the basis of an unacceptable proportion of neonates born asymptomatic, defined as less
than 60%. This was derived from the results of a historical comparator group, and acceptable evidence
of a positive effect of valaciclovir, defined as the proportion of neonates born asymptomatic of more
than 80%. In these conditions, valaciclovir was judged to have a positive effect if at least 31 out of 43
cases were asymptomatic at birth. The 41 included women (43 fetuses) were treated for a median of 89
days and valaciclovir since 34 neonates were asymptomatic. In this study the maternal clinical and
laboratory tolerances to high dosage acyclovir were excellent and no adverse effect was observed in the
neonates. Moreover, despite the burden of taking 16 tablets throughout the day, cumulative adherence
to treatment exceeded 90%. The main limitation of this study was that it was not randomised. It
therefore remains difficult to assess definitively the respective roles of true antiviral effect and a
placebo effect to explain the positive effect of treatment demonstrated in this setting.

Summary

Recent epidemiological data from longitudinal adequately powered studies confirm that CMV
infection is one of the most prevalent congenital diseases but also provide good evidence that the
contribution of maternal secondary infections to the burden of congenital infections is higher than
previously reported. Unfortunately, there are still no validated tools to diagnose these maternal sec-
ondary infections. Conversely, there is a good level of evidence from numerous studies based on IgG,
IgM and IgG avidity results that the diagnosis of maternal primary infection as well as the estimation of
its timing are feasible with a good sensitivity and specificity. In recent years, a number of studies from
different groups have demonstrated that establishing the prognosis of an infected fetus based on the
combination of imaging and laboratory parameters is feasible, even as early aa the time of prenatal
diagnosis, with very high negative predictive values. This is crucial for counselling women with an
infected fetus and it opens up opportunities to target a group at high risk for progressive damage that
could be considered for potential fetal treatment. Valaciclovir treatment in women carrying a
moderately infected fetus is safe but the level of evidence for its efficacy is not yet high and need
confirmation from a randomized study. It is also important to underline that in the absence of
screening policies a large number of symptomatic fetuses will continue to be undiagnosed because
systematic ultrasound during pregnancy has low sensitivity for detecting those cases.

Practice points

- Diagnosis of primary maternal infection is accurate when based on maternal IgG, IgM and IgG
avidity serology (level III)
- Diagnosis of fetal infection is accurate when based on CMV PCR in amniotic fluid collected
after 20 weeks of gestation and 6 weeks after the presume maternal primary infection. (level
III)
- Prognosis of an infected fetus is accurately when based on a combination of imaging and fetal
laboratory parameters. (level II)

Please cite this article in press as: Leruez-Ville M, Ville Y, Fetal cytomegalovirus infection, Best Practice &
Research Clinical Obstetrics and Gynaecology (2016), http://dx.doi.org/10.1016/j.bpobgyn.2016.10.005
 M. Leruez-Ville, Y. Ville / Best Practice & Research Clinical Obstetrics and Gynaecology xxx (2016) 1e11 9

Research agenda

- Reliable diagnostic tools for maternal secondary infection

- Effectiveness of hyperimmune globulins to prevent transmission in randomized control trial
- Effectiveness of antiviral therapy to treat fetal infection in a randomized control trial

Acknowledgements

I hereby confirm that I have no conflict of interest to declare in relation with the topic of congenital
CMV infection nor with the pharmaceutical or imaging industry that takes no part in the design or
funding of my research.

References

*[1] 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 Oct;17(5):355e63.
*[2] Mussi-Pinhata MM, Yamamoto AY, Moura Brito RM, et al. Birth prevalence and natural history of congenital cytomeg-
alovirus infection in a highly seroimmune population. Clin Infect Dis 2009 Aug 15;49(4):522e8.
*[3] Wang C, Zhang X, Bialek S, et al. Attribution of congenital cytomegalovirus infection to primary versus non-primary
maternal infection. Clin Infect Dis 2011 Jan 15;52(2):e11e3.
*[4] Foulon I, Naessens A, Foulon W, et al. 10-year prospective study of sensorineural hearing loss in children with congenital
cytomegalovirus infection. J Pediatr 2008 Jul;153(1):84e8.
*[5] de Vries JJC, van Zwet EW, Dekker FW, et al. The apparent paradox of maternal seropositivity as a risk factor for congenital
cytomegalovirus infection: a population-based prediction model. Rev Med Virol 2013 Jul;23(4):241e9.
*[6] Fowler KB, Stagno S, Pass RF, et al. The outcome of congenital cytomegalovirus infection in relation to maternal antibody
status. N Engl J Med 1992 Mar 5;326(10):663e7.
*[7] Ross SA, Fowler KB, Ashrith G, et al. Hearing loss in children with congenital cytomegalovirus infection born to mothers
with preexisting immunity. J Pediatr 2006 Mar;148(3):332e6.
*[8] Townsend CL, Forsgren M, Ahlfors K, et al. Long-term outcomes of congenital cytomegalovirus infection in Sweden and
the United Kingdom. Clin Infect Dis Off Publ Infect Dis Soc Am 2013 May;56(9):1232e9.
*[9] Sijmons S, Thys K, Mbong Ngwese M, et al. High-throughput analysis of human cytomegalovirus genome diversity
highlights the widespread occurrence of gene-disrupting mutations and pervasive recombination. J Virol 2015 May 13.
*[10] Boppana SB, Rivera LB, Fowler KB, et al. Intrauterine transmission of cytomegalovirus to infants of women with pre-
conceptional immunity. N Engl J Med 2001 May 3;344(18):1366e71.
[11] Revello MG, Furione M, Zavattoni M, et al. Human cytomegalovirus (HCMV) DNAemia in the mother at amniocentesis as a
risk factor for iatrogenic HCMV infection of the fetus. J Infect Dis 2008 Feb 15;197(4):593e6.
[12] Grangeot-Keros L, Mayaux MJ, Lebon P, et al. Value of cytomegalovirus (CMV) IgG avidity index for the diagnosis of
primary CMV infection in pregnant women. J Infect Dis 1997 Apr;175(4):944e6.
[13] Mace
M, Sissoeff L, Rudent A, et al. A serological testing algorithm for the diagnosis of primary CMV infection in pregnant
women. Prenat Diagn 2004 Nov;24(11):861e3.
[14] Munro SC, Hall B, Whybin LR, et al. Diagnosis of and screening for cytomegalovirus infection in pregnant women. J Clin
Microbiol 2005 Sep;43(9):4713e8.
[15] Guerra B, Simonazzi G, Banfi A, et al. Impact of diagnostic and confirmatory tests and prenatal counseling on the rate of
pregnancy termination among women with positive cytomegalovirus immunoglobulin M antibody titers. Am J Obstet
Gynecol 2007 Mar;196(3). 221.e1-6.
[16] Revello MG, Genini E, Gorini G, et al. Comparative evaluation of eight commercial human cytomegalovirus IgG avidity
assays. J Clin Virol 2010 Aug;48(4):255e9.
[17] Leruez-Ville M, Sellier Y, Salomon LJ, et al. Prediction of fetal infection in cases with cytomegalovirus immunoglobulin M
in the first trimester of pregnancy: a retrospective cohort. Clin Infect Dis 2013 May;56(10):1428e35.
[18] Sellier Y, Guilleminot T, Ville Y, et al. Comparison of the LIAISON(®) CMV IgG Avidity II and the VIDAS(®) CMV IgG Avidity
II assays for the diagnosis of primary infection in pregnant women. J Clin Virol Off Publ Pan Am Soc Clin Virol 2015 Nov;
72:46e8.
[19] Vauloup-Fellous C, Lazzarotto T, Revello MG, et al. Clinical evaluation of the Roche Elecsys CMV IgG avidity assay. Eur J
Clin Microbiol Infect Dis Off Publ Eur Soc Clin Microbiol 2014 Aug;33(8):1365e9.
[20] Bode
us M, Hubinont C, Bernard P, et al. Prenatal diagnosis of human cytomegalovirus by culture and polymerase chain
reaction: 98 pregnancies leading to congenital infection. Prenat Diagn 1999 Apr;19(4):314e7.
[21] Lazzarotto T, Varani S, Spezzacatena P, et al. Maternal IgG avidity and IgM detected by blot as diagnostic tools to identify
pregnant women at risk of transmitting cytomegalovirus. Viral Immunol 2000;13(1):137e41.
[22] Bode
us M, Van Ranst M, Bernard P, et al. Anticytomegalovirus IgG avidity in pregnancy: a 2-year prospective study. Fetal
Diagn Ther 2002 Dec;17(6):362e6.
[23] Liesnard C, Donner C, Brancart F, et al. Prenatal diagnosis of congenital cytomegalovirus infection: prospective study of
237 pregnancies at risk. Obstet Gynecol 2000 Jun;95(6 Pt 1):881e8.

Please cite this article in press as: Leruez-Ville M, Ville Y, Fetal cytomegalovirus infection, Best Practice &
Research Clinical Obstetrics and Gynaecology (2016), http://dx.doi.org/10.1016/j.bpobgyn.2016.10.005
10 M. Leruez-Ville, Y. Ville / Best Practice & Research Clinical Obstetrics and Gynaecology xxx (2016) 1e11

[24] Guerra B, Lazzarotto T, Quarta S, et al. Prenatal diagnosis of symptomatic congenital cytomegalovirus infection. Am J
Obstet Gynecol 2000 Aug;183(2):476e82.
[25] Lazzarotto T, Varani S, Guerra B, et al. Prenatal indicators of congenital cytomegalovirus infection. J Pediatr 2000 Jul;
137(1):90e5.
[26] Enders G, Ba €der U, Lindemann L, et al. Prenatal diagnosis of congenital cytomegalovirus infection in 189 pregnancies with
known outcome. Prenat Diagn 2001 May;21(5):362e77.
[27] Bilavsky E, Pardo J, Attias J, et al. Clinical implications for children born with congenital cytomegalovirus infection
following a negative amniocentesis. Clin Infect Dis 2016 Apr 24.
[28] Gindes L, Teperberg-Oikawa M, Sherman D, et al. Congenital cytomegalovirus infection following primary maternal
infection in the third trimester. BJOG 2008 Jun;115(7):830e5.
[29] Picone O, Vauloup-Fellous C, Cordier AG, et al. A series of 238 cytomegalovirus primary infections during pregnancy:
description and outcome. Prenat Diagn 2013 Aug;33(8):751e8.
[30] Drose JA, Dennis MA, Thickman D. Infection in utero: US findings in 19 cases. Radiology 1991 Feb;178(2):369e74.
[31] MacGregor SN, Tamura R, Sabbagha R, et al. Isolated hyperechoic fetal bowel: significance and implications for man-
agement. Am J Obstet Gynecol 1995 Oct;173(4):1254e8.
[32] De
chelotte PJ, Mulliez NM, Bouvier RJ, et al. Pseudo-meconium ileus due to cytomegalovirus infection: a report of three
cases. Pediatr Pathol Affil Int Paediatr Pathol Assoc 1992 Feb;12(1):73e82.
[33] Chaoui R, Zodan-Marin T, Wisser J. Marked splenomegaly in fetal cytomegalovirus infection: detection supported by
three-dimensional power Doppler ultrasound. Ultrasound Obstet Gynecol Off J Int Soc Ultrasound Obstet Gynecol 2002
Sep;20(3):299e302.
[34] Stein B, Bromley B, Michlewitz H, et al. Fetal liver calcifications: sonographic appearance and postnatal outcome.
Radiology 1995 Nov;197(2):489e92.
[35] Noyola DE, Demmler GJ, Nelson CT, et al. Early predictors of neurodevelopmental outcome in symptomatic congenital
cytomegalovirus infection. J Pediatr 2001 Mar;138(3):325e31.
[36] Holzgreve W, Feil R, Louwen F, et al. Prenatal diagnosis and management of fetal hydrocephaly and lissencephaly. Childs
Nerv Syst ChNS Off J Int Soc Pediatr Neurosurg 1993 Nov;9(7):408e12.
[37] Nigro G, La Torre R, Sali E, et al. Intraventricular haemorrhage in a fetus with cerebral cytomegalovirus infection. Prenat
Diagn 2002 Jul;22(7):558e61.
[38] Malinger G, Lev D, Zahalka N, et al. Fetal cytomegalovirus infection of the brain: the spectrum of sonographic findings.
AJNR Am J Neuroradiol 2003 Jan;24(1):28e32.
[39] Estroff JA, Parad RB, Teele RL, et al. Echogenic vessels in the fetal thalami and basal ganglia associated with cytomega-
lovirus infection. J Ultrasound Med Off J Am Inst Ultrasound Med 1992 Dec;11(12):686e8.
[40] Benoist G, Salomon LJ, Mohlo M, et al. Cytomegalovirus-related fetal brain lesions: comparison between targeted ul-
trasound examination and magnetic resonance imaging. Ultrasound Obstet Gynecol 2008 Dec;32(7):900e5.
[41] Picone O, Simon I, Benachi A, et al. Comparison between ultrasound and magnetic resonance imaging in assessment of
fetal cytomegalovirus infection. Prenat Diagn 2008 Aug;28(8):753e8.
[42] Lipitz S, Hoffmann C, Feldman B, et al. Value of prenatal ultrasound and magnetic resonance imaging in assessment of
congenital primary cytomegalovirus infection. Ultrasound Obstet Gynecol Off J Int Soc Ultrasound Obstet Gynecol 2010
Dec;36(6):709e17.
[43] Gaytant MA, Steegers EAP, Semmekrot BA, et al. Congenital cytomegalovirus infection: review of the epidemiology and
outcome. Obstet Gynecol Surv 2002 Apr;57(4):245e56.
[44] Benoist G, Salomon LJ, Jacquemard F, et al. The prognostic value of ultrasound abnormalities and biological parameters in
blood of fetuses infected with cytomegalovirus. BJOG Int J Obstet Gynaecol 2008 Jun;115(7):823e9.
[45] Guerra B, Simonazzi G, Puccetti C, et al. Ultrasound prediction of symptomatic congenital cytomegalovirus infection. Am J
Obstet Gynecol 2008 Apr;198(4). 380.e1-7.
[46] Farkas N, Hoffmann C, Ben-Sira L, et al. Does normal fetal brain ultrasound predict normal neurodevelopmental outcome
in congenital cytomegalovirus infection? Prenat Diagn 2011 Apr;31(4):360e6.
[47] Lipitz S, Yinon Y, Malinger G, et al. Risk of cytomegalovirus-associated sequelae in relation to time of infection and
findings on prenatal imaging. Ultrasound Obstet Gynecol Off J Int Soc Ultrasound Obstet Gynecol 2013 May;41(5):
508e14.
[48] Leruez-Ville M, Stirnemann J, Sellier Y, et al. Feasibility of predicting the outcome of fetal infection with cytomegalovirus
at the time of prenatal diagnosis. Am J Obstet Gynecol 2016 Apr 8.
[49] Revello MG, Zavattoni M, Furione M, et al. Quantification of human cytomegalovirus DNA in amniotic fluid of mothers of
congenitally infected fetuses. J Clin Microbiol 1999 Oct;37(10):3350e2.
[50] Gouarin S, Gault E, Vabret A, et al. Real-time PCR quantification of human cytomegalovirus DNA in amniotic fluid samples
from mothers with primary infection. J Clin Microbiol 2002 May;40(5):1767e72.
[51] Picone O, Costa J-M, Leruez-Ville M, et al. Cytomegalovirus (CMV) glycoprotein B genotype and CMV DNA load in the
amniotic fluid of infected fetuses. Prenat Diagn 2004 Dec 15;24(12):1001e6.
[52] Goegebuer T, Van Meensel B, Beuselinck K, et al. Clinical predictive value of real-time PCR quantification of human
cytomegalovirus DNA in amniotic fluid samples. J Clin Microbiol 2009 Mar;47(3):660e5.
[53] Lanari M, Lazzarotto T, Venturi V, et al. Neonatal cytomegalovirus blood load and risk of sequelae in symptomatic and
asymptomatic congenitally infected newborns. Pediatrics 2006 Jan;117(1):e76e83.
[54] Ross SA, Novak Z, Fowler KB, et al. Cytomegalovirus blood viral load and hearing loss in young children with congenital
infection. Pediatr Infect Dis J 2009 Jul;28(7):588e92.
[55] Fabbri E, Revello MG, Furione M, et al. Prognostic markers of symptomatic congenital human cytomegalovirus infection
in fetal blood. BJOG Int J Obstet Gynaecol 2011 Mar;118(4):448e56.
[56] Nigro G, Adler SP, La Torre R, et al. Congenital Cytomegalovirus Collaborating Group. Passive immunization during
pregnancy for congenital cytomegalovirus infection. N Engl J Med 2005 Sep 29;353(13):1350e62.
[57] Revello MG, Lazzarotto T, Guerra B, et al. A randomized trial of hyperimmune globulin to prevent congenital cytomeg-
alovirus. N Engl J Med 2014 Apr 3;370(14):1316e26.

Please cite this article in press as: Leruez-Ville M, Ville Y, Fetal cytomegalovirus infection, Best Practice &
Research Clinical Obstetrics and Gynaecology (2016), http://dx.doi.org/10.1016/j.bpobgyn.2016.10.005
 M. Leruez-Ville, Y. Ville / Best Practice & Research Clinical Obstetrics and Gynaecology xxx (2016) 1e11 11

[58] Tyms AS, Scamans EM, Naim HM. The in vitro activity of acyclovir and related compounds against cytomegalovirus
infections. J Antimicrob Chemother 1981 Jul;8(1):65e72.
[59] Wutzler P, Thust R. Genetic risks of antiviral nucleoside analoguesea survey. Antiviral Res 2001 Feb;49(2):55e74.
[60] Stone KM, Reiff-Eldridge R, White AD, et al. Pregnancy outcomes following systemic prenatal acyclovir exposure: con-
clusions from the international acyclovir pregnancy registry, 1984-1999. Birt Defects Res A Clin Mol Teratol 2004 Apr;
70(4):201e7.
[61] Pasternak B, Hviid A. Use of acyclovir, valacyclovir, and famciclovir in the first trimester of pregnancy and the risk of birth
defects. JAMA J Am Med Assoc 2010 Aug 25;304(8):859e66.
[62] Jacquemard F, Yamamoto M, Costa J-M, et al. Maternal administration of valaciclovir in symptomatic intrauterine
cytomegalovirus infection. BJOG Int J Obstet Gynaecol 2007 Sep;114(9):1113e21.
res L, et al. In utero treatment of congenital cytomegalovirus infection with valacyclovir in
[63] Leruez-Ville M, Ghout I, Bussie
a multicenter, open-label, phase II study. Am J Obstet Gynecol 2016 Apr 13.

Please cite this article in press as: Leruez-Ville M, Ville Y, Fetal cytomegalovirus infection, Best Practice &
Research Clinical Obstetrics and Gynaecology (2016), http://dx.doi.org/10.1016/j.bpobgyn.2016.10.005