Keywords
cytomegalovirus fetal infection - nonprimary infection - lissencephaly
Palavras-chave
infeção fetal por citomegalovírus - infeção não primária - lisencefalia
Introduction
Cytomegalovirus (CMV) is the most common congenital viral infection, with a birth
prevalence of 0,2 to 2,2%,[1] causing hearing and visual impairment and delay of psychomotor development.[1]
Primary CMV infection during pregnancy has an incidence of 1 to 4%,[2] and the global risk of transplacental infection is ∼ 40%.[3]
[4] The risk of fetal infection increases with the gestational age[1]
[2]
[5] (according to a study involving 508 pregnancies, vertical CMV transmission rate
by gestational age at the time of maternal infection was 0% in the pregestation group,
4.6% in the periconception group, 34.8, 42 and 58.6% respectively in the first, second
and third trimester groups).[3] Severe disease may affect fetuses infected during the first trimester as well as
the second, but rarely affects fetuses infected later on in pregnancy.[3]
[5] Also, congenital infection impairs placental development and functions, and should
be considered as an underlying cause of intrauterine growth restriction (IUGR), regardless
of virus transmission to the fetus.[6]
Ten to 15% of congenitally infected newborns are overtly symptomatic at birth and
will present neurodevelopmental damage within the first three years.[1] From those, 20 to 30% die, mostly of disseminated intravascular coagulation, hepatic
dysfunction, or bacterial superinfection; severe neurologic morbidity occurs in 50
to 60% of survivors.[1]
[5]
[7] Of the asymptomatic infants at birth (85 to 90%), 5 to 15% will develop long-term
neurodevelopmental morbidity, such as sensorial hearing loss, delay of psychomotor
development and visual impairment.[1]
The criteria necessary to recommend universal screening for CMV infection in pregnant
women haven't yet been met, and are not recommended by the Centers for Disease Control
and Prevention or the American College of Obstetricians and Gynecologists.[2]
[8] Prenatal diagnosis is mainly performed when US abnormalities are observed.[4]
Nonprimary maternal CMV infection (reactivation or reinfection with a different strain)
occurs more rarely, with less probability of fetal infection (1,5% in a study with
543 cases of suspected maternal nonprimary infection)[9] and only 0.2 to 2% of newborns will develop a symptomatic disease,[6] which indicates a minor fetal affectation in these cases.
This case draws attention to the gravity of a nonprimary CMV infection.
Case Description
The patient is 37 years old, with irrelevant personal and family medical histories,
and in her fourth spontaneous pregnancy. The patient has a history of two spontaneous
abortions and a term vaginal delivery, which resulted in a healthy child. The spouse
is healthy; the couple is not consanguineous.
Preconceptional analysis indicated maternal immunity to Rubella, Toxoplasmosis and
CMV. First trimester routine analysis at the 11th week of pregnancy indicated CMV
IgG and IgM of respectively 848.9 Ul/mL (positive threshold ≥ 1) and 0.379 Ul/mL (negative
threshold < 0,8). The first trimester US did not show fetal anomalies, and the combined
prenatal screening for chromosomal abnormalities was negative.
At 20 weeks of gestation, US revealed cerebral abnormalities: thin and hyperechogenic
cerebral cortex with prominent lateral ventricles, bilateral periventricular hyperechogenicities,
cerebellar vermis hypoplasia and absent corpus callosum ([Fig. 1]). Amniocentesis was performed and the karyotype was normal (46, XY).
Fig. 1 (A) Ultrasound at 20 weeks of gestation – thin and hyperechogenic cerebral cortex
with prominent lateral ventricles and bilateral periventricular hyperechogenicities.
(B) Ultrasound at 20 weeks of gestation – cerebellar vermis hypoplasia (arrow).
An extended serology study was then performed, and it showed no immunity to Herpes
simplex virus type 2 and immunity to Toxoplasmosis, Herpes simplex virus type 1, Parvovirus
and CMV. However, CMV IgG title increased almost three times since the first trimester
analysis, and CMV IgM kept negative (IgG 2695.0 Ul/mL and IgM 0.333 Ul/mL) – these
serologic tests were performed in the same laboratory as the first trimester tests.
IgG avidity testing was high.
When asked about flu–like symptoms, the patient mentions an episode during the second
trimester of pregnancy; however, as it was mild, it was difficult to precise the time
of it.
Magnetic resonance imaging (MRI) at 22 weeks of gestation confirmed US findings and
raised the hypothesis of fetal infection: “[…] enlargement of the cerebrospinal fluid
circulation spaces at the expense of marked reduction in the thickness of the cerebral
parenchyma of both hemispheres of lissencephalic appearance [...] suggesting serious
sequelae of brain infection (group TORCH? other?) rather than primary malformative
injury of the central nervous system” ([Fig. 2]).
Fig. 2 Magnetic resonance imaging at 22 weeks of gestation with lissencephalic appearance.
Outcome and Follow-up
The pregnancy was terminated at 23 weeks.
At this time, a new sample of amniotic fluid was collected to perform PCR for CMV
DNA. The result was negative. Pathologic examination study of fetal and placental
tissues revealed cerebral ventricular dilation with necro-hemorrhagic and calcified
periventricular lesions, necro-hemorrhagic lesions also at the level of the basal
ganglia, hippocampus, periventricular regions of brainstem and cerebellar hemispheres.
There were areas of four-layer polymicrogyria and frequent cellular viral inclusions
(cytomegalic cells), particularly in the germinal matrix and near the necrotic lesions.
These cells tested positive for CMV immunohistochemistry ([Figs. 3], [4] and [5]). Immunohistochemical detection of CMV DNA was found not only in the fetus brain,
but also in his liver, heart and in the placenta.
Fig. 3 Macroscopical findings – coronal section at the level of the hippocampus showing
dilation of lateral ventricles, extensive white matter periventricular lesions with
focal necro-hemorrhagic lesions at this level and also at the level of the amygdala/hippocampus
and thalamus.
Fig. 4 Histopathological findings showing four-layer polymicrogyria (HE). Magnification,
50x.
Fig. 5 (A) Histopathological findings – neurons of germinal matrix with some viral inclusions
(arrow), with the typical aspect of cytomegalic cells (HE). Magnification, 630x. (B)
Histopathological findings – cells positive for CMV immunohistochemistry (some showing
colocation with the viral inclusions). Magnification, 100x.
Discussion
All the investigation and outcomes in this case suggest a probable CMV secondary infection,
which can be due to a reactivation or reinfection with a different strain. According
to the maternal serology and the morphologic US performed, the maternal CMV secondary
infection must have occurred between 11 and 20 weeks of pregnancy.
The preferred approach for diagnosing fetal infection is PCR testing for DNA virus
in the amniotic fluid,[1]
[3]
[4] which has a sensitivity of 75–100%.[4] Gestational age at amniocentesis and the time elapsed since the suspicion of infection
are critical factors for the sensitivity of the method.[1]
[3]
[4]
The PCR for CMV DNA negative result was indeed a false negative. Although this method
is recognized as the gold standard for prenatal diagnosis of fetal CMV infection,
the gestational age and the time lag between maternal infection and amniocentesis
is a critical factor influencing sensitivity. To achieve the highest sensitivity,
amniocentesis should be performed after 21 weeks of gestation and at least 6 to 7
weeks after the onset of maternal infection reflecting the time it takes for placental
infection, transmission to the fetus, viral replication in the fetal kidney, and excretion
into amniotic fluid.[1]
[3]
[8] In spite of having fulfilled the first prerequisite, the time lag might not have
been achieved. Nevertheless, we were reaching the legal limit for terminating the
pregnancy, and even without a prenatal diagnosis, the US findings foretold a poor
outcome for the offspring.
The idea that nonprimary CMV maternal infection will not result in severe fetal sequelae
and congenital disease has been proved different in several published cases.[5]
[10]
[11] The preexisting maternal CMV seropositivity substantially reduces, but not eliminates,
the risk of fetal infection and affectation; therefore, the use of prenatal invasive
methods for the detection of fetal infection in nonprimary maternal infections may
be justified.
Neurons migrate from the areas where they are born to the areas where they will settle
into their proper neural circuits. Neuronal migration starts in the second month of
gestation and is controlled by a complex assortment of chemical guides and signals.
When these signals are absent or incorrect, neurons do not end up where they belong,
resulting in a primary malformative injury called lissencephaly. The lack of gyral
development after a defective neuronal migration in the developing brain can be associated
with the agenesis of the corpus callosum or cerebellar hypoplasia, and is also due
to some genetic syndromes.[12]
Neuropathological studies of brains infected prenatally by CMV show evidence of lissencephaly,
polymicrogyria, cerebellar hypoplasia and ventricular dilation in consequence of loss
of volume (atrophy),[7]
[12] findings suggestive of the severity and risk of neurological and motor sequelae.
Still, a normal US examination cannot exclude neonatal or long-term morbidity.[1]
[13] Mild ventriculomegaly, cerebellar vermis hypoplasia and lissencephaly can also be
associated with extra central nervous system abnormalities, chromosomal aberration
and other fetal infections like rubella and toxoplasmosis.[13] In this case, no other abnormalities were found beyond the fetal central nervous
system; the karyotype was normal and the other fetal infections were excluded.
There is insufficient evidence to support the use of passive immunization to prevent
congenital infection,[2] and the only known means of reducing the risk of congenital CMV is by reducing exposure
to the virus through hygiene measures.[2]
[5]
Therapies for treatment of congenital CMV are in the experimental stage. In an uncontrolled
study published in 2005, administration of CMV-specific hyperimmune globulin to pregnant
women with primary CMV infection significantly reduced the rate of intrauterine transmission,
from 40 to 16%.[14] A 2014 randomized placebo-controlled phase 2 trial of virus-specific hyperimmune
globulin for the prevention of congenital CMV infection, involving 124 women with
primary CMV infection from 5 to 26 weeks of gestation, showed that the treatment with
hyperimmune globulin did not significantly modify the course of primary CMV infection
during pregnancy.[15] Other phase 3 studies are under way to further our understanding of the efficacy
and safety of hyperimmune globulin administration as a means of preventing congenital
CMV infection.
Although it is still not recommended, our institution offers pregnant women systematic
screening for CMV infection, as well as education about hygienic measures. However,
we do not provide any therapy for pregnant women with suspicious or confirmed CMV
primary or nonprimary infections.
Nevertheless, the US and MRI findings in this case suggested the presence of severe
advanced disease and high risk of long-term neurodevelopmental impairment, so the
option of pregnancy termination was obviously raised.
This article is about a rare case of nonprimary maternal CMV infection during pregnancy,
with vertical transmission, resulting in severe fetal affectation. The preexisting
maternal CMV seropositivity substantially reduces, but not eliminates, the risk of
fetal infection and affectation. This article highlights the importance of clinical
suspicion and the difficulty of prenatal confirmation of a CMV infection.
