33 CME REVIEWARTICLE Volume 65, Number 11

OBSTETRICAL AND GYNECOLOGICAL SURVEY

Copyright © 2011
by Lippincott Williams & Wilkins

CHIEF EDITOR’S NOTE: This article is part of a series of continuing education activities in this Journal through which a total
of 36 AMA/PRA Category 1 CreditsTM can be earned in 2011. Instructions for how CME credits can be earned appear on the
last page of the Table of Contents.

Screening, Diagnosis, and Management of

Cytomegalovirus Infection in Pregnancy
Yoav Yinon, MD,* Dan Farine, MD, FRCSC,†
and Mark H. Yudin, MD, MSc, FRCSC†‡
*Staff Perinatologist, Fetal Medicine Unit, Department of Obstetrics and Gynecology, Sheba Medical Center,
Tel Hashomer, Tel Aviv University, Israel and Department of Obstetrics & Gynaecology, Mount Sinai
Hospital, University of Toronto, Toronto, Ontario, Canada; †Professor and Head of Maternal-Fetal
Medicine, Department of Obstetrics & Gynaecology, Mount Sinai Hospital, University of Toronto, Toronto,
Ontario, Canada; and ‡Assistant Professor and Attending Staff Physician, Department of Obstetrics and
Gynecology, St. Michael’s Hospital, University of Toronto, Toronto, Ontario, Canada

Congenital cytomegalovirus (CMV) is the most common intrauterine infection and the leading
infectious cause of sensorineural hearing loss and mental retardation. This article reviews the
issues that relate to the diagnosis and management of this disease, detailing the points that led to
the recent published guidelines by the Society of Obstetricians and Gynaecologists of Canada.
A MEDLINE/Cochrane search of CMV infection, pregnancy, and prenatal diagnosis found 195
studies between 1980 and 2010. Of these, we examined 59 relevant studies. The probability of
intrauterine transmission following primary infection is 30% to 40%, but only 1% after secondary
infection. About 10% to 15% of congenitally infected infants will have symptoms at birth, and 20%
to 30% of them will die, whereas 5% to 15% of the asymptomatic infected neonates will develop
sequelae later. Children with congenital CMV infection following first trimester infection are more
likely to have central nervous system sequelae, whereas infection acquired in the third trimester
has a high rate of intrauterine transmission but a favorable outcome.
The prenatal diagnosis of fetal CMV infection should be based on amniocentesis performed 7
weeks after the presumed time of infection and after 21 weeks of gestation. Sonographic findings
often imply poor prognosis, but their absence does not guarantee a normal outcome. The value of
quantitative determination of CMV DNA in the amniotic fluid is not yet confirmed. The effectiveness
of prenatal therapy for fetal CMV is not yet proven, although CMV-specific hyperimmune globulin
may be beneficial. Routine serologic screening of pregnant women or newborns has never been
recommended by any public health authority.
Target Audience: Obstetricians & Gynecologists, Family Physicians
Learning Objectives: After completion of this educational activity, the obstetrician/gynecologist should
be better able to evaluate the principles of prenatal diagnosis of congenital CMV infection so doctors will
be familiar with the tests and procedures needed, in order to reach a diagnosis of congenital CMV; to
assess the natural history and outcome of congenital CMV infection enabling obstetricians to counsel
prenatally pregnant women with CMV; and to analyze the prognostic markers for fetal CMV, so managing
physicians will be able to predict more accurately the outcomes of fetuses infected by CMV.

Unless otherwise noted below, each faculty’s and staff’s spouse/life partner (if any) has nothing to disclose.
The authors have disclosed that they have no financial relationships with or interests in any commercial companies pertaining to this
educational activity. Each faculty’s and staff’s spouse/life partner (if any) has nothing to disclose.
The faculty and staff in a position to control the content of this CME activity have disclosed that they have no financial relationships
with, or financial interests in, any commercial companies pertaining to this educational activity.
Correspondence requests to: Mark H. Yudin, MD, St. Michael’s Hospital, 30 Bond St, Toronto, Ontario M5B 1W8, Canada. E-mail:
yudinm@smh.toronto.on.ca.
www.obgynsurvey.com | 736
 Management of Cytomegalovirus Infection in Pregnancy Y CME Review Article 737

Cytomegalovirus (CMV) occurs in 0.2% to 2.2% a review of the diagnostic and prognostic implications
of all live births and is the most common cause of of CMV in pregnancy, and an examination of recent
intrauterine infection and the leading infectious cause developments in treatment and disease prevention.
of sensorineural hearing loss and mental retardation
(1,2). CMV is the largest known member of the
human herpes virus family. These viruses are large METHODS
enveloped DNA viruses, sharing the biologic prop- We performed a systematic review of the literature
erties of latency and reactivation (3). regarding prenatal diagnosis, outcomes, and treat-
Although CMV is found throughout all geographic ment of congenital CMV infection. Studies were
locations and in all socioeconomic groups, it is more identified by electronic searches of the MEDLINE
widespread in developing countries and in areas of database and the Cochrane Library for the time pe-
lower socioeconomic conditions. Factors that have riod of 1980 to 2010. The keywords used were con-
been associated with seropositivity include lower so- genital CMV infection, pregnancy, and prenatal
cioeconomic status, maternal age more than 30 years, diagnosis. The reference lists of relevant articles re-
nonwhite race, lower level of education, and close trieved by the searches were also reviewed. Reports
contact with young children (4). Transmission of included in our review were limited to those written
CMV occurs from person to person and requires in English. Studies were eligible if the diagnosis of
intimate contact with infected excretions, such as congenital CMV infection was based on CMV cul-
saliva, urine, or other body fluids (5,6). Seropreva- ture or polymerase chain reaction (PCR) of the am-
lence is associated with socioeconomic status, and in niotic fluid and confirmed by neonatal urine culture
the United States and Western Europe, rates in young for CMV after birth. The literature search yielded
women of childbearing age range from 40% for 195 articles with potential relevance. After screening
women of middle to upper socioeconomic status to these articles on the basis of title and abstract, 59
83% for women of lower socioeconomic status (3). articles were identified and used for our review.
Most healthy individuals who acquire CMV have
no symptoms or long-term consequences. A small
proportion has mild symptoms of malaise, fever, Fetal CMV Infection and Postnatal Outcome
myalgia, and lymphadenopathy (3).
Primary Infection
An important distinction must be made between
primary and secondary infection. Primary infection Maternal viremia, placental infection, and hematog-
occurs in a seronegative person who has never been enous dissemination to the fetus is the most likely
infected before. Following primary infection, the vi- sequence of events leading to congenital CMV infec-
rus persists in a latent state. Secondary, or recurrent, tion after primary maternal infection (11). Transmission
infection occurs when an individual with a history of can occur after either primary or secondary infection,
primary infection has a reactivation of the latent virus but the likelihood is much greater after primary infec-
(5). Secondary infection can also occur due to a tion, with a probability of 30% to 40% (1,12).
second infection with a different strain of the virus. The burden of disease for congenitally infected
Molecular analysis of viral isolates is required to infants is high, with 10% to15% having symptoms at
distinguish between these 2 types of secondary birth, including intrauterine growth restriction, mi-
infection (3,7,8). crocephaly, hepatosplenomegaly, petechiae, jaun-
In 1% to 4% of seronegative pregnancies there is dice, chorioretinitis, thrombocytopenia, and anemia.
seroconversion, with women of low socioeconomic sta- Of infants who are symptomatic at birth, 20% to 30%
tus or poor personal hygiene experiencing higher rates will die, and 90% of the symptomatic survivors will
(9). Immunosuppression during pregnancy may con- have late complications (12–14). Despite 85% to
tribute to the increase in the incidence of primary or 90% of congenitally infected infants showing no
secondary CMV infections in pregnant women (4). signs or symptoms at birth, late sequelae appear in
In April 2010, the Society of Obstetricians and 5% to 15%. These include sensorineural hearing loss,
Gynaecologists of Canada published guidelines for pre- delay of psychomotor development, and visual im-
natal diagnosis and management of the disease in pairment (15,16) (Fig. 1).
pregnancy (10) (available at: http://www.sogc.org/ Data regarding transmission rates according to ges-
jogc/abstracts/full/201004_SOGCClinicalPractice tational age are not consistent. While primary CMV
Guidelines_1.pdf). The purpose of this article is to infection acquired either before or around conception
elaborate on the reason for these published guidelines in carries the lowest risk of transmission (17), CMV
738 Obstetrical and Gynecological Survey

Primary Secondary Infection

Infection
As noted above, fetal infection may occur follow-
ing both primary and recurrent maternal infection,
but the difference in the incidence of transmission is
10-15% 30-40% 85-90%
Symptomatic remarkable. During recurrent infection, virus repli-
Fetal Asymptomatic
at Birth Infection cation occurs in the presence of both humoral and
cell-mediated immune responses. As a result, viremia
occurs as a rule only in primary infection and, there-
5-15% fore, the transmission rate is about 40% following
20-30% 90% Late Complications primary infection and only about 1% in cases of
Mortality Complications recurrent maternal infection (11). Fowler et al found
Fig. 1. Fetal outcome after primary CMV infection. evidence of intrauterine CMV infection in 3% of
infants born to 604 mothers who were seronegative at
the beginning of pregnancy, and in 1% of the infants
transmission rates appear to increase with advancing
born to 2857 mothers who were seropositive before
stages of pregnancy. Daiminger et al demonstrated
pregnancy (23). These results show that maternal
transmission rates of 30%, 39%, and 58% following
preconceptional immunity against CMV gives rela-
primary infection at gestational weeks 6 to 20, 18 to
tively good protection to the fetus, but a small pro-
22, and 20 to 38, respectively (18). Revello and
portion may still become infected. Yamamoto et al
Gerna reported transmission rates of 45.4%, 45.6%,
have recently shown that maternal reinfection by new
and 78.6% following primary infection in the first,
strains of CMV is a major source of congenital
second, and third trimesters, respectively (11). Bo-
infection in CMV seroimmune women. Seroconver-
deus et al have recently reported that the rate of
sion to new CMV strains during pregnancy was
transmission increases gradually during gestation,
observed in 17.5% of women delivering infected
based on 524 seroconversions. The transmission rate
infants compared with only 4.6% in control mothers
was 34.5% during the first trimester, 44.1% during
of uninfected infants (24).
the second trimester, and 73.3% when seroconver-
The traditional belief has been that most children
sion occurred during the third trimester (19).
with congenital CMV born to mothers who had sec-
Despite the higher transmission rate with maternal
ondary CMV infection are asymptomatic at birth,
infection occurring later in pregnancy, the rate of
and fewer than 10% of them develop sequelae,
sequelae in infected offspring appears to be lower.
mainly sensorineural hearing loss and chorioretinitis
Liesnard et al studied 55 cases of congenital CMV
(16,25). There is, however, increasing evidence in
infection and found 10/38 (26%) cases infected be-
recent years that the incidence of symptomatic infec-
fore 20 weeks of gestation had severe disease, com-
tion in infants born to immune mothers following
pared with only 1/16 (6.2%) case infected after 20
secondary infection may be higher than previously
weeks (20). Pass et al found sensorineural hearing
thought (25–27). Gaytant et al summarized 6 studies
loss in 8/34 (24%) of first trimester cases compared
that reported on the type of maternal infection in
with 1/40 (2.5%) in the later infection group (21),
relation to the outcome of congenital infection. There
and Daiminger’s series included 18 cases of infection
were 50 cases of congenital CMV following second-
occurring after 20 weeks’ gestation, of which none
ary infection, in which 45 (90%) were asymptomatic
had evidence of congenital disease (18). Gindes et al
at birth. One (2%) developed minor symptoms (mod-
have recently reported on 28 women with primary
erate psychomotor retardation, behavioral problems,
CMV infection acquired after 25 weeks of gestation,
clumsiness), and 4 (9%) developed major symptoms
of whom 21 (75%) had evidence of fetal infection but
(severe mental retardation, hearing loss). Of the 5
none of the live infected newborns had congenital
infants who were symptomatic at birth, 2 (40%)
CMV disease (22).
developed major symptoms (3).
In summary, the research shows that children with
congenital CMV infection following first trimester
maternal infection are more likely to have central
Prenatal Diagnosis
nervous system sequelae, whereas CMV infection
acquired during the third trimester is associated with There are 2 important components in prenatal di-
a high rate of intrauterine transmission but a more agnosis of congenital CMV infection. The first is
favorable outcome for the infant. distinguishing between maternal primary and sec-
 Management of Cytomegalovirus Infection in Pregnancy Y CME Review Article 739

ondary infection based on serologic testing (16). The nancy. Although ultrasound may be used, most ul-
second is identifying whether fetal infection is pres- trasound findings of CMV infection can also be seen
ent in women with proven CMV infection by using with other infections or conditions affecting the fe-
both noninvasive (ultrasound examination) and inva- tus. In addition, these findings are only seen in less
sive (amniocentesis) prenatal testing (16). than 25% of infected fetuses (32). Ultrasound find-
ings of fetal CMV infection consist of fetal growth
Diagnosis of Maternal Infection restriction, cerebral ventriculomegaly, ascites, intra-
cranial calcifications, abnormality of amniotic fluid
The diagnosis of primary CMV infection is clearly volume (usually oligohydramnios), microcephaly,
established when seroconversion can be documented. hyperechogenic bowel, hydrops fetalis, pleural effu-
However, this is only possible with screening pro- sion, and liver calcifications (32–34).
grams that identify seronegative women and then Compared with ultrasound findings, CMV isola-
follow them prospectively and re-test them to iden- tion from amniotic fluid has a much higher sensitiv-
tify seroconversion (25). When the immune status ity and specificity and is considered the gold standard
before pregnancy is unknown, determination of pri- for prenatal diagnosis of fetal CMV infection
mary CMV infection should be based on detection of (13,25,35). Replication of the virus in the fetal kid-
the specific IgM antibody. Unfortunately, this ap- ney to produce sufficient quantity to be identified in
proach has limitations, since IgM can be detected in amniotic fluid occurs only after 5 to 7 weeks from the
10% of recurrent infections (28) and, unlike with onset of fetal infection. In addition, gestational age at
many other infections, can be detected for months after time of amniocentesis has been shown to be an
primary infection (29). Therefore, in addition to pri- additional variable affecting sensitivity. Liesnard et
mary infection during pregnancy, CMV IgM positivity al demonstrated a sensitivity of 30% if the first
may indicate either remote primary infection acquired amniotic fluid sample was taken before 21 weeks of
before pregnancy or recurrent infection (20). gestation, increasing to 71% thereafter (20). Therefore,
Because of these difficulties with interpreting the
amniocentesis should be performed at least 7 weeks
serology, the IgG avidity assay can be a useful tool to
after the onset of maternal infection and after 21 weeks
assist in distinguishing primary infection from past or
of gestation (11,13,20). Amniocenteses performed too
recurrent infection and can assist in dating the time of
close to the onset of maternal infection carry a substan-
infection (25,30). It is known that in the first few
months after infection, virus-specific IgG of low tial risk of false negative results (36–38).
avidity appears. Over time, the IgG antibody shows The diagnosis of fetal CMV infection should be
increasingly higher avidity. This high avidity is de- based on culture and/or PCR performed on amniotic
tectable only with remote or recurrent CMV infection fluid samples. Some culture techniques allow detec-
(25). An avidity index ⬍30% strongly suggests a tion of the virus 16 to 24 hours after amniotic fluid
primary infection of less than 3 months (30). collection (11,39,40).
Using serology, a diagnosis of primary CMV in- With the advent of PCR, which allows the ampli-
fection during pregnancy is documented by either fication to a detectable level of minute amounts of
seroconversion (the appearance of CMV-specific viral DNA present in the amniotic fluid, the sensitiv-
IgG antibody in a previously seronegative woman) or ity of prenatal diagnosis of fetal infection has in-
detection of specific IgM antibody associated with creased. Revello and Gerna have reported that the
low IgG avidity. Recurrent infection is diagnosed in sensitivity, specificity, and positive and negative pre-
women with detectable specific IgG antibodies with- dictive values of DNA detection in amniotic fluid
out IgM antibodies before pregnancy and a signifi- obtained in a series of 102 pregnant women were
cant increase of IgG antibody titer with or without 90.2%, 100%, 100%, and 90.4%, respectively (11).
the presence of specific IgM antibodies and high IgG Despite the use of a very sensitive technique such as
avidity (31). PCR, it is reasonable to assume that a delay in
intrauterine transmission of the infection may repre-
sent a major obstacle to achieving 100% sensitivity
Diagnosis of Fetal Infection (25). However, several studies have indicated that the
Once maternal infection has been documented, it is detection of even small amounts of viral DNA in the
important to determine whether fetal infection has amniotic fluid correlate with congenital infection at
also occurred, as this will help to guide the need for birth, explaining the 100% specificity of this test
further evaluation and surveillance during preg- (41,42).
740 Obstetrical and Gynecological Survey

Another tool available for diagnosis is the use of women with primary infection. Twenty-three (15%)
fetal CMV IgM. However, this is not recommended of 154 fetuses with congenital infection had abnor-
because cordocentesis is associated with risk, and mal sonographic findings, of whom 18 (78%) were
there is poor sensitivity, as it is often late in preg- symptomatic, whereas among the 131 infected fe-
nancy that many fetuses develop specific IgM (9,37). tuses without abnormal sonographic findings, 68
Lipitz et al studied 63 pregnant women with primary fetuses/neonates (52%) were classified as symptom-
CMV infection, in whom fetal diagnosis was made atic, resulting in a negative predictive value of only
by amniocentesis and fetal blood sampling or amnio- 48% (43). Therefore, although ultrasound is a valu-
centesis only. Thirteen of 22 patients with evidence able tool in evaluating a fetus with CMV infection,
of fetal infection by amniocentesis underwent cordo- its limitations should be mentioned when counseling
centesis, and 10 of the 13 (77%) showed positive patients.
IgM results in the fetal blood. No case of positive Recent progress in the field of magnetic resonance
fetal serum IgM with negative amniotic fluid culture imaging (MRI) has contributed to the development of
or PCR was recorded, indicating that the information detailed fetal imaging. Two recent studies have eval-
yielded by cordocentesis did not increase the ability uated the contribution of MRI to the diagnosis of
to accurately diagnose intrauterine infection (37). fetal brain abnormalities in CMV-infected fetuses
In a large series of 237 women who had primary (44,45). The first series included 11 cases with no
CMV infection studied by amniocentesis with or ultrasound features, in which MRI always confirmed
without cordocentesis, the best sensitivity and 100% the absence of visceral or cerebral anomalies. How-
specificity were achieved by PCR done on amniotic ever, in 13 cases with extracerebral features without
fluid sampled after 21 weeks of gestation with a cerebral abnormalities at ultrasound, MRI revealed
mean interval of 7 weeks between maternal infection cerebral anomalies in 6 cases (46%) and therefore
and amniocentesis (20). modified the prognosis. Termination of pregnancy
Amniocentesis may be used in cases of either was performed in 5 of these 6 fetuses (encephalitis
primary or secondary maternal CMV infection. Al- was found at fetopathological examination in all 5
though the risk of fetal infection is lower with sec- cases), and 1 child is mentally retarded (44). Benoist
ondary infection, it may occasionally result in severe et al have shown that the best positive predictive
sequelae. Therefore, the risk/benefit ratio of perform- value (88.9%) for prenatal diagnosis of cerebral le-
ing this invasive diagnostic test must be considered sions was obtained with a combination of abnormal
carefully in cases of secondary infection. ultrasound and MRI findings, but the MRI did not
improve the negative predictive value obtained by
normal ultrasound findings (45).
Prognostic Markers of CMV Disease
Another marker which has been studied as a prog-
Although prenatal testing can identify infection, nostic factor is CMV viral load in amniotic fluid.
the presence of the virus in the amniotic fluid does Guerra et al reported that a CMV viral load ⬎103
not reliably predict fetal outcomes. Detection of genome equivalents was 100% predictive of fetal
sonographic abnormalities may aid in determining infection, whereas a level ⬎105 was predictive of
fetal prognosis, but there is no guarantee of a normal symptomatic CMV disease (46). Similarly, other
fetus/infant with no sequelae if these findings are studies have also shown significantly higher CMV
absent. DNA load values in amniotic fluid samples in the
Lipitz et al have studied the outcomes of 50 preg- group of symptomatic fetuses, as opposed to the
nancies with documented intrauterine CMV infec- asymptomatic group. However, there was great over-
tion, in which 17 pregnancies (18 fetuses) continued lap between viral load values in the symptomatic and
to term. Four fetuses had neurologic abnormalities, asymptomatic groups, and other variables such as ges-
of which 3 had normal prenatal ultrasound findings, tational age at time of amniocentesis and time elapsed
giving a risk of 19% (3 of 16) for postnatal neuro- since maternal infection were found to influence viral
logic abnormalities. These problems included hear- load irrespective of fetal outcome (41,42). Therefore,
ing loss, chorioretinitis, and developmental delay, more study is needed on the prognostic value of CMV
even when there were no prenatal sonographic DNA viral load levels.
abnormalities (32). Determination of a few parameters in the fetal
Guerra et al have recently reported on the effec- blood, such as CMV-specific IgM, viral load, as well
tiveness of ultrasound in the antenatal prediction of as assessment of biochemical and hematological pa-
symptomatic congenital CMV infection in 600 rameters, might also assist in predicting fetal out-
 Management of Cytomegalovirus Infection in Pregnancy Y CME Review Article 741

come. A significant correlation between high levels In addition to prenatal therapy, there is also an
of virus-specific IgM and adverse fetal outcome has option of treatment after birth for symptomatic in-
been reported (47,48). Moreover, all virologic pa- fants. Some evidence has suggested limited benefit
rameters tested to determine viral load in fetal blood with ganciclovir treatment of neonates with symp-
were found to be higher in fetuses with abnormalities tomatic congenital CMV infection. A study carried
compared to fetuses with normal findings (48). These out on a group of 47 infants with congenital infec-
data might indicate that congenitally infected fetuses tion, who were treated with ganciclovir for 6 weeks,
with normal biochemical, hematological, and ultra- showed hearing improvement in 5 of 30 infants
sound findings and low viral load in blood, together (16%) after 6 months of follow-up. The most com-
with low IgM antibody, may have a more favorable mon side effects were neutropenia and elevation of
outcome (48). liver enzymes (51). In another more recent study,
Thus, fetal blood sampling, although not sensitive ganciclovir was given within the first month at 12
enough to justify its use for the detection of intra- mg/kg/d intravenously for 6 weeks (52). Of the 42
uterine infection, may provide important prognostic infants followed, evaluation at 6 months and 1 year
information. More studies are needed to validate showed significantly less hearing deterioration in
these fetal blood measures as prognostic markers treated infants compared with control infants (52).
and, therefore, fetal blood sampling should not be Similarly, Michaels et al found no progression of
routinely performed in the prenatal evaluation of hearing loss at a median age of 2 in 9 treated children
congenital CMV infection. with congenital CMV, 5 of whom had hearing loss
before therapy was started. However, there was no
evidence of improvement in neurodevelopmental
Prenatal Treatment and Prevention of
sequelae (53).
Congenital CMV Infection
The best prevention for congenital CMV infection
Although there are several tools available for the is primary prevention, and this could be accom-
prenatal diagnosis of congenital CMV infection, plished with a vaccine. Vaccination of seronegative
there is no effective treatment to offer once a diag- women of childbearing age could prevent the occur-
nosis has been made. The use of CMV-specific hy- rence of primary CMV infection during pregna-
perimmune globulin for treatment was evaluated ncy. In a phase 2, placebo-controlled, randomized,
recently in 157 pregnant women with primary CMV double-blind trial, Pass et al have evaluated the effi-
infection (49). cacy of the CMV glycoprotein B vaccine compared
After amniocentesis, CMV was found in the am- with a placebo. The vaccine group was more likely to
niotic fluid of 45 women with primary infection more remain uninfected during a 42-month period than the
than 6 weeks before enrolment. Of the 31 women placebo group (P ⫽ 0.002). However, the sample
who elected to receive intravenous treatment with size of this study was too small to determine the
CMV hyperimmune globulin (200 U per kilogram of efficacy of the vaccine in preventing congenital in-
the mother’s body weight), only one (1/31) had an fection (54). Therefore, until the efficacy of this
infant with clinical CMV disease at birth. Fifteen of vaccine is established, recommendations for avoid-
the 31 had abnormal ultrasound findings. Of the 14 ing infection revolve mostly around personal hygiene
women who declined treatment, 7 (7/14) had infants practices, such as thorough handwashing after diaper
who were symptomatic at delivery. changes, and avoiding intimate contact with salivary
In the prevention group, 37 women received hy- secretions and urine from young children (55).
perimmune globulin and 6 (16%) of them had infants To determine if protective behavior prevents
with congenital CMV infection. In comparison, 19 of child-to-mother transmission of CMV during preg-
47 women (40%) who did not receive hyperimmune nancy, Adler et al studied 166 seronegative mothers
globulin had infected infants. No adverse effects of with a child ⬍36 months of age attending a daycare
the hyperimmune globulin were observed (49). facility (56). Mothers, either pregnant or attempting
These results are provocative, and this may be the pregnancy, were randomly assigned to either a con-
first effective antenatal treatment that is available. trol or intervention group, in which the mothers re-
However, it is important to note that this was not a ceived instructions for handwashing, glove use, and
randomized controlled trial and further study is nec- avoiding intimate contact with their children. The pro-
essary. Prenatal administration of ganciclovir into the portion of women seroconverting was the same in both
umbilical vein has also been reported but its value in the intervention and control groups, at 7.8% of women.
improving the prognosis is not well established (50). However, for 41 women attempting pregnancy at
742 Obstetrical and Gynecological Survey

enrolment with a child shedding CMV, 10 of 24 be- available, research should be focused on identifying
came infected compared with only 1 of 17 women who reliable prognostic markers and on validating the
were already pregnant at enrolment (56). According to promising reports on the effectiveness of CMV-
this study, intervention before pregnancy was found to specific hyperimmune globulin in treating the disease
be ineffective, but intervention for pregnant women antenatally.
may be effective, as pregnant women may be more
motivated than nonpregnant women to adhere to rec- REFERENCES
ommendations to protect their unborn children (57).
1. Stagno S, Pass RF, Cloud G, et al. Primary cytomegalovirus
infection in pregnancy. Incidence, transmission to fetus, and
The Issue of Screening clinical outcome. JAMA 1986;256:1904–1908.
2. Pultoo A, Jankee H, Meetoo G, et al. Detection of cytomega-
There is no consensus on the issue of screening for lovirus in urine of hearing-impaired and mentally retarded
children by PCR and cell culture. J Commun Dis 2000;32:101–
CMV by serology. No public health authority has 108.
ever recommended routine serologic screening for 3. Gaytant MA, Steegers EA, Seminmekrot BA, et al. Congenital
pregnant women (25). The time for screening would cytomegalovirus infection: review of the epidemiology and
outcome. Obstet Gynecol Surv 2002;57:245–256.
be at the beginning of pregnancy or in advance of 4. Ornoy A, Diav-Citrin O. Fetal effects of primary and secondary
pregnancy. In the case of a seronegative woman, cytomegalovirus infection in pregnancy. Reprod Toxicol 2006;
serologic testing for CMV would need to be repeated 21:399–409.
5. Hanshaw JB. Cytomegalovirus infections. Pediatr Rev 1995;
during pregnancy to rule out seroconversion. 16:43–48; quiz 49.
Arguments against screening include the fact that 6. Stagno S, Pass RF, Dworsky ME, et al. Maternal cytomega-
there is no effective vaccine and no treatment that has lovirus infection and perinatal transmission. Clin Obstet Gy-
necol 1982;25:563–576.
been proven effective. For infections with effective 7. Alford CA, Stagno S, Pass RF, et al. Congenital and perinatal
vaccines, screening is usually done either before or at cytomegalovirus infections, Rev Infect Dis 1990;12:S745–
the beginning of pregnancy. As there is no effective S753.
8. Daniel Y, Gull I, Peyser MR, et al. Congenital cytomegalovirus
prenatal treatment, the choices with regard to an infection. Eur J Obstet Gynecol Reprod Biol 1995;63:7–16.
infected fetus are to terminate the pregnancy or to 9. Hagay ZJ, Biran G, Ornoy A, et al. Congenital cytomegalovirus
observe the fetus until delivery. On the other hand, infection: a long-standing problem still seeking a solution.
Am J Obstet Gynecol 1996;174:241–245.
with prenatal testing women can be educated about 10. Yinon Y, Farine D, Yudin MH, et al. Cytomegalovirus infection
behavior and prevention with respect to seronegative in pregnancy. J Obstet Gynaecol Can 2010;32:348–354.
women (58). 11. Revello MG, Gerna G. Pathogenesis and prenatal diagnosis of
human cytomegalovirus infection. J Clin Virol 2004;29:71–83.
Finally, routine antibody testing, especially if done 12. Raynor BD. Cytomegalovirus infection in pregnancy. Semin
before pregnancy, may help to differentiate between Perinatol 1993;17:394–402.
primary and secondary infection in cases of suspected 13. Nigro G, Mazzocco M, Anceschi MM, et al. Prenatal diagnosis
of fetal cytomegalovirus infection after primary or recurrent
CMV infection during pregnancy (4). Naessens et al maternal infection. Obstet Gynecol 1999;94:909–914.
evaluated a CMV screening program in which sero- 14. Pass RF. Cytomegalovirus infection. Pediatr Rev 2002;23:
logical testing was performed at the first prenatal 163–170.
15. Boppana SB, Pass RF, Britt WJ, et al. Symptomatic congen-
visit. With this program, screening detected 82% of ital cytomegalovirus infection: neonatal morbidity and mortal-
all congenital CMV infections (59). However, as ity. Pediatr Infect Dis J 1992;11:93–99.
long as there is no available vaccine or effective 16. Lazzarotto T, Varani S, Guerra B, et al. Prenatal indicators
of congenital cytomegalovirus infection. J Pediatr 2000;
prenatal treatment, it will be difficult to justify sero- 137:90–95.
logical screening for CMV in terms of cost- 17. Revello MG, Zavattoni M, Furione M, et al. Diagnosis and
effectiveness. Therefore, routine serologic testing of outcome of preconceptional and periconceptional primary hu-
man cytomegalovirus infections. J Infect Dis 2002;186:553–
all pregnant women for CMV, to identify primary 557.
infection in pregnancy, is not recommended. 18. Daiminger A, Bader U, Enders G. Pre- and periconceptional
primary cytomegalovirus infection: risk of vertical transmis-
sion and congenital disease. BJOG 2005;112:166–172.
CONCLUSION 19. Bodeus M, Zech F, Hubinont C, et al. Human cytomegalovirus
in utero transmission: Follow-up of 524 maternal seroconver-
Congenital CMV infection is a substantial concern sions. J Clin Virol 2010;47:201–202.
20. Liesnard C, Donner C, Brancart F, et al. Prenatal diagnosis of
in pregnant women. Following a diagnosis of, pri- congenital cytomegalovirus infection: prospective study of
mary maternal CMV infection, fetal infection may be 237 pregnancies at risk. Obstet Gynecol 2000;95:881–888.
confirmed by amniocentesis. However, our ability to 21. Pass RF, Fowler KB, Boppana SB, et al. Congenital cytomeg-
alovirus infection following first trimester maternal infection:
predict the outcome of an infected fetus is quite symptoms at birth and outcome. J Clin Virol 2006;35:216–
limited. Until an effective and safe vaccine becomes 220.
 Management of Cytomegalovirus Infection in Pregnancy Y CME Review Article 743

22. Gindes L, Teperberg-Oikawa M, Sherman D, et al. Congenital ples from mothers with primary infection. J Clin Microbiol
cytomegalovirus infection following primary maternal infection 2002;40:1767–1772.
in the third trimester. BJOG 2008;115:830–835. 42. Revello MG, Zavattoni M, Furione M, et al. Quantification of
23. Fowler KB, Stagno S, Pass RF. Maternal immunity and pre- human cytomegalovirus DNA in amniotic fluid of mothers of
vention of congenital cytomegalovirus infection. JAMA 2003; congenitally infected fetuses. J Clin Microbiol 1999;37:3350–
289:1008–1011. 3352.
24. Yamamoto AY, Mussi-Pinhata MM, Boppana SB, et al. 43. Guerra B, Simonazzi G, Puccetti C, et al. Ultrasound predic-
Human cytomegalovirus reinfection is associated with tion of symptomatic congenital cytomegalovirus infection.
intrauterine transmission in a highly cytomegalovirus- Am J Obstet Gynecol 2008;198:380.e1–380.e7.
immune maternal population. Am J Obstet Gynecol 2010; 44. Picone O, Simon I, Benachi A, et al. Comparison between
202:e1–e8. ultrasound and magnetic resonance imaging in assessment of
25. Revello MG, Gerna G. Diagnosis and management of human fetal cytomegalovirus infection. Prenat Diagn 2008;28:753–
cytomegalovirus infection in the mother, fetus, and newborn 758.
infant. Clin Microbiol Rev 2002;15:680–715. 45. Benoist G, Salomon LJ, Mohlo M, et al. Cytomegalovirus-
26. Boppana SB, Fowler KB, Britt WJ, et al. Symptomatic con- related fetal brain lesions: comparison between targeted
genital cytomegalovirus infection in infants born to mothers ultrasound examination and magnetic resonance imaging.
with preexisting immunity to cytomegalovirus. Pediatrics Ultrasound Obstet Gynecol 2008;32:900–905.
1999;104:55–60. 46. Guerra B, Lazzarotto T, Quarta S, et al. Prenatal diagnosis of
27. Gaytant MA, Rours GI, Steegers EA, et al. Congenital cyto- symptomatic congenital cytomegalovirus infection. Am J Ob-
megalovirus infection after recurrent infection: case reports stet Gynecol 2000;183:476–482.
and review of the literature. Eur J Pediatr 2003;162:248–253. 47. Enders G, Bader U, Lindemann L, et al. Prenatal diagnosis of
28. Griffiths PD, Stagno S, Pass RF, et al. Infection with cytomeg- congenital cytomegalovirus infection in 189 pregnancies with
alovirus during pregnancy: specific IgM antibodies as a known outcome. Prenat Diagn 2001;21:362–377.
marker of recent primary infection. J Infect Dis 1982;145:647– 48. Revello MG, Zavattoni M, Baldanti F, et al. Diagnostic and
653. prognostic value of human cytomegalovirus load and IgM
29. Drew WL. Diagnosis of cytomegalovirus infection. Rev Infect antibody in blood of congenitally infected newborns. J Clin
Dis 1988;10(suppl 3):S468–S476. Virol 1999;14:57–66.
30. Grangeot-Keros L, Mayaux MJ, Lebon P, et al. Value of cy- 49. Nigro G, Adler SP, La Torre R, et al. Passive immunization
tomegalovirus (CMV) IgG avidity index for the diagnosis of during pregnancy for congenital cytomegalovirus infection.
primary CMV infection in pregnant women. J Infect Dis 1997; N Engl J Med 2005;353:1350–1362.
175:944–946. 50. Revello MG, Percivalle E, Baldanti F, et al. Prenatal treatment
31. Yinon Y, Yagel S, Tepperberg-Dikawa M, et al. Prenatal diag- of congenital human cytomegalovirus infection by fetal intra-
nosis and outcome of congenital cytomegalovirus infection in vascular administration of ganciclovir. Clin Diagn Virol 1993;
twin pregnancies. BJOG 2006;113:295–300. 1:61–67.
32. Lipitz S, Achiron R, Zalel Y, et al. Outcome of pregnancies 51. Whitley RJ, Cloud G, Gruber W, et al; National Institute of
with vertical transmission of primary cytomegalovirus infec- Allergy and Infectious Diseases Collaborative Antiviral Study
tion. Obstet Gynecol 2002;100:428–433. Group. Ganciclovir treatment of symptomatic congenital cy-
33. Crino JP. Ultrasound and fetal diagnosis of perinatal infection. tomegalovirus infection: results of a phase II study. J Infect
Clin Obstet Gynecol 1999;42:71–80; quiz 174–175. Dis 1997;175:1080–1086.
34. Malinger G, Lev D, Zahalka N, et al. Fetal cytomegalovirus 52. Kimberlin DW, Lin CY, Sanchez PJ, et al. Effect of ganciclovir
infection of the brain: the spectrum of sonographic findings. therapy on hearing in symptomatic congenital cytomegalovi-
Am J Neuroradiol 2003;24:28–32. rus disease involving the central nervous system: a random-
35. Hohlfeld P, Vial Y, Maillard-Brignon C, et al. Cytomegalovirus ized, controlled trial. J Pediatr 2003;143:16–25.
fetal infection: prenatal diagnosis. Obstet Gynecol 1991;78: 53. Michaels MG, Greenberg DP, Sabo DL, et al. Treatment of
615–618. children with congenital cytomegalovirus infection with gan-
36. Bodeus M, Hubinont C, Bernard P, et al. Prenatal diagnosis of ciclovir. Pediatr Infect Dis J 2003;22:504–509.
human cytomegalovirus by culture and polymerase chain re- 54. Pass RF, Zhang C, Evans A, et al. Vaccine prevention of
action: 98 pregnancies leading to congenital infection. Prenat maternal cytomegalovirus infection. N Engl J Med 2009;360:
Diagn 1999;19:314–317. 1191–1199.
37. Lipitz S, Yagel S, Shalev E, et al. Prenatal diagnosis of fetal 55. Adler SP, Finney JW, Manganello AM, et al. Prevention of
primary cytomegalovirus infection. Obstet Gynecol 1997;89: child-to-mother transmission of cytomegalovirus by changing
763–767. behaviors: a randomized controlled trial. Pediatr Infect Dis J
38. Nicolini U, Kustermann A, Tassis B, et al. Prenatal diagnosis of 1996;15:240–246.
congenital human cytomegalovirus infection. Prenat Diagn 56. Adler SP, Finney JW, Manganello AM, et al. Prevention of
1994;14:903–906. child-to-mother transmission of cytomegalovirus among
39. Gleaves CA, Smith TF, Shuster EA, et al. Rapid detection of pregnant women. J Pediatr 2004;145:485–491.
cytomegalovirus in MRC-5 cells inoculated with urine speci- 57. Adler SP, Nigro G, Pereira L. Recent advances in the preven-
mens by using low-speed centrifugation and monoclonal an- tion and treatment of congenital cytomegalovirus infections.
tibody to an early antigen. J Clin Microbiol 1984;19:917–919. Semin Perinatol 2007;31:10–18.
40. Lazzarotto T, Varani S, Gabrielli L, et al. New advances in the 58. Demmler GJ. Screening for congenital cytomegalovirus infec-
diagnosis of congenital cytomegalovirus infection. Intervirol- tion: a tapestry of controversies. J Pediatr 2005;146:162–164.
ogy 1999;42:390–397. 59. Naessens A, Casteels A, Decatte L, et al. A serologic strategy
41. Gouarin S, Gault E, Vabret A, et al. Real-time PCR quantifi- for detecting neonates at risk for congenital cytomegalovirus
cation of human cytomegalovirus DNA in amniotic fluid sam- infection. J Pediatr 2005;146:194–197.