Accelerat ing t he world's research.

Congenital cytomegalovirus
infection: recent advances in the
diagnosis of maternal infection
Brunella Guerra
Human Immunology

Cite this paper Downloaded from Academia.edu 

Get the citation in MLA, APA, or Chicago styles

Related papers Download a PDF Pack of t he best relat ed papers 

Prenat al diagnosis of congenit al cyt omegalovirus infect ion

Pat rick Hohlfeld

Updat e on t he prevent ion, diagnosis and management of cyt omegalovirus infect ion during pregnancy
Brunella Guerra

Guidelines on CMV congenit al infect ion

F. Bot et -mussons, Oriol Coll
Congenital Cytomegalovirus Infection:
Recent Advances in the Diagnosis of
Maternal Infection
Tiziana Lazzarotto, Liliana Gabrielli, Marcello Lanari,
Brunella Guerra, Tatiana Bellucci, Monica Sassi, and
Maria Paola Landini
ABSTRACT: In most European countries, pregnant
women are tested for cytomegalovirus (CMV) during the
first trimester of pregnancy. Within the last 5 years,
European laboratories have made significant progress in
solving diagnostic problems linked to infection in preg-
nancy. With advances in CMV serology, the presence of
anti-CMV immunoglobulin (Ig)M detected by a screen-
ing test such as enzyme immunoassay, can be confirmed
by blot, identifying pregnant women undergoing an ac-
tive or recent infection. Furthermore, primary infections
that were proven if a seroconversion was observed or
suspected in the presence of IgM, can now be readily
diagnosed by disclosing the presence of anti-CMV low
avidity in IgM-positive mothers, greatly reducing the
number of women who should be considered at risk of
transmitting the virus. Virologic maternal tests are not
ABBREVIATIONS
CMV cytomegalovirus
EIA enzyme immunoassay
GE genome equivalents
NPV negative predictive value
INTRODUCTION
Cytomegalovirus (CMV) is the most common cause of
intrauterine infection, affecting 0.3–2% of live-born in-
fants [1]. Transmission of CMV from mother to fetus can
occur throughout gestation, but infection during the first
16 weeks of pregnancy has been associated with a higher
From the Department of Clinical and Experimental Medicine (Section of
Microbiology) (T.L., L.G., T.B., M.S., M.P.L.) and Department of Ob-
stetrics and Gynaecology (B.G.), St. Orsola Malpighi General Hospital,
University of Bologna, Bologna, Italy; Department of Paediatrics and
Neonatology, La Scaletta Hospital (M.L.), Imola-Bologna, Italy.
Address reprint requests to: Dr. T. Lazzarotto, Section of Microbiology,
Policlinico S. Orsola-Malpighi, Via Massarenti n. 9, 40138 Bologna, Italy;
Tel: ⫹39.051.346306; Fax: ⫹39.051.307397; E-mail: titti@med.
unibo.it.
Received August 12, 2003; revised January 15, 2004; accepted February
3, 2004.
Human Immunology 65, 410⫺415 (2004)
© American Society for Histocompatibility and Immunogenetics, 2004
Published by Elsevier Inc.
enough to diagnose a recent primary maternal CMV in-
fection and the detection or quantification of CMV in
maternal blood does not seem to be associated with a
higher risk for fetal infection. A cohort of 1520 pregnant
women considered at risk of transmitting the virus were
followed in a longitudinal study at the University of
Bologna. Women were identified as part of routine CMV
screening in several Italian regions and were IgM-positive
for CMV. We documented IgG seroconversion in 83
women and 1437 were IgM-positive by commercial
kit. Human Immunology 65, 410⫺415 (2004). ©
American Society for Histocompatibility and Immunoge-
netics, 2004. Published by Elsevier Inc.
KEYWORDS: Cytomegalovirus; congenital infection;
antibody avidity; IgM blot
PCR polymerase chain reaction
PMNL polymorphonuclear leukocyte
PPV positive predictive value
qPCR quantitative PCR
incidence of damage [2]. The rate of vertical transmission
was 0.2–2.2% in previously seropositive mothers under-
going recurrent infection during pregnancy and 20 –
50% in pregnant women with primary infection [2, 3].
The diagnosis of primary CMV infection is straight-
forward if seroconversion to CMV-specific antibodies is
detected. However, because documentation of CMV se-
roconversion is rare because women are not routinely
screened for CMV antibodies before gestation, the detec-
tion of CMV immunoglobulin (Ig)M has been used as a
marker of active or recent CMV infection. Because CMV-
specific IgM can be produced during nonprimary CMV
infections, related to the persistence of IgM antibody, a
CMV IgM-positive result alone does not accurately pre-
dict the risk fetal infection [4, 5].
0198-8859/04/$–see front matter
doi:10.1016/j.humimm.2004.02.006
Congenital Cytomegalovirus Infection
Among recent advances in CMV serology, the deter-
mination of anti-CMV IgG avidity has gained diagnostic
importance in identifying primary infection [4 –7].
However, this does not solve the problem because only
20 –50% of infected mothers will give birth to infected
infants [2].
The aim of this article is to study different approaches
and recent advances in the development of new serologic
and virologic tests used to screen pregnant women for
CMV infection and to reliably identify those women who
should be offered prenatal diagnosis.
MATERIALS AND METHODS
In several Italian regions, pregnant women are advised to
undergo screening for IgG and IgM anti-CMV before 10
weeks’ gestation. Seronegative women are rechecked at
the end of the fourth month of pregnancy. Women with
test results showing seroconversion or IgM positivity are
sent to a referral center for further diagnostic investiga-
tion. The screening is performed with commercially
available kits for both anti-CMV IgG and IgM.
Between 1994 and 2003, a cohort of 1520 pregnant
women considered at risk of transmitting the virus were
followed in a longitudinal study at our unit. At the time
of recruitment, the patients were in the first and second
trimester of pregnancy, except 71 who were in the third
trimester (range 7–34 weeks’ gestation, median 15).
The first step of our diagnostic protocol was based on
redetermination of CMV-specific IgG and IgM by en-
zyme immunoassay (EIA) [8] on IgM confirmation by
blot [9] combined with determination of the avidity of
anti-CMV IgG [5]. This IgM blot has a high sensitivity
(100%) and specificity (99%) [9].
Women who were found to be either CMV-seroneg-
ative (for both IgG and IgM) or IgG-positive with a high
avidity index of anti-CMV IgG- and IgM-negative were
defined as “uninfected/latently infected.” Those who had
anti-CMV IgG of low avidity combined with true IgM or
who seroconverted to CMV IgG positivity were classified
as undergoing primary infection. Those with anti-CMV
IgG of high avidity and blot-confirmed IgM positivity
were classified as having a recurrent infection. Finally,
doubtful serologic results were classified as “undefined.”
The second step of our diagnostic protocol was to
establish whether detection of CMV in maternal blood is
a helpful tool to identify all primarily infected women
who will give birth to an infected newborn. From our
cohort of 1520 pregnant women, 94 blood samples were
obtained from 94 women with a definite diagnosis of
primary CMV infection.
Of 94 polymorphonuclear leukocyte (PMNL) samples,
94 were examined for antigenemia [10], 66 for polymer-
ase chain reaction (PCR), and 67 for quantitative PCR
(qPCR) [10, 11].
411
With respect to the outcome, fetuses and newborns
were classified as uninfected or infected. Newborns were
classified on the basis of virus isolation from urine sam-
pled within the first week of life; aborted fetuses were
classified on the basis of macroscopic and histologic
tissue examination (brain, heart, lungs, liver, spleen,
pancreas, kidneys) [12]. Infected newborns were further
classified as having symptomatic or asymptomatic infec-
tion if they displayed more than one of the following
clinical features: jaundice, hepatosplenomegaly, petechial
or purpuric rush, chorioretinitis, microcephaly, cerebral
calcifications, or low birth weight. Infants were consid-
ered to have adverse outcome if they displayed more than
one of the following signs during subsequent evalua-
tions: microcephaly, developmental delay, vision or hear-
ing loss, seizures, or abnormal muscle tone (cerebral
palsy) [2].
Histologic examinations were carried out by the pa-
thology departments of the hospitals where the preg-
nancy termination was performed. The entire study was
carried out following the ethical rules of the S. Orsola
General Hospital (Bologna, Italy). The statistical analysis
was performed using Fisher’s one-tailed exact test [13].
RESULTS
Over the last 10 years (1994 –2003), serum samples
taken from 1520 women referred to our unit for sus-
pected CMV infection were serologically tested. We
documented IgG seroconversion in 83 women, and 1437
were IgM positive by different commercial kits.
As shown in Table 1, no active infection was observed
in 789 cases, a diagnosis of recurrent infection was made
in 272 cases, a diagnosis of primary infection was made
in 399 (26.2%) women, whereas no indication of the
type of infection was obtained in 60 cases (“undefined”).
Among the 1520 pregnant women, 789 were uncon-
firmed IgM and all samples had a high avidity IgG
index. A total of 710 newborns were checked and no
CMV congenital infection was identified. Of 731 women
with a positive blot profile considered at risk, 272 had a
high avidity anti-CMV IgG and 4 (1.7%) infected in-
fants were detected (3 asymptomatic, 1 severely symp-
tomatic).
From the data presented in Table 1, 316 serum sam-
ples were with both anti-CMV IgM and low avidity IgG,
indicating a primary CMV infection. Among 316 preg-
nant women considered at risk, 286 fetuses/infants were
examined and 87 of them were infected (40 symptom-
atic, 47 asymptomatic). Of 40 (14.7%) cases with symp-
toms, 24 (8.4%) developed late sequelae (data not
shown).
Of the 83 pregnant women who seroconverted, con-
genital infections were observed in 30 cases; 9 (10.8%)
412 T. Lazzarotto et al.

TABLE 1 Results of diagnosis of maternal CMV infection by serologic tests and pregnancy outcomes
Total no. No. of No. of specimens with No. of fetuses/newborns
of specimens
specimens pos by Low Moderate High Infected
Group (%) IgM blot AI AI AI Examined (%) Lost

No active infection 789 (51.9%) 0 0 0 789 710 0 79

Recurrent infection 272 (17.9%) 272 0 0 272 234 4 (1.7%) 38
Primary infection 316 (20.8%) 316 316 0 0 286 87 (30.4%) 30
True IgM
combined with
low avidity
Primary infection 83 (5.5%) 83 83 0 0 83 30 (36.1%) 0
Seroconversion
Undefined 60 (3.9%) 60 0 60 0 50 2 (4.0%) 10
infection
Total 1520 731 399 60 1061 1363 123 157

Abbreviations: CMV ⫽ cytomegalovirus; Ig ⫽ immunoglobulin; pos ⫽ positive; AI ⫽ avidity-IgG index.

fetuses/infants symptomatic at birth and 6 (7.2%) devel-
oped late sequelae.
Finally, only two infected newborns were found
among the 60 pregnant women with undefined CMV
infection (one severely symptomatic, one asymptomatic).
In these cases, the anti-CMV IgM blot profile combined
with moderate avidity index did not differentiate pri-
mary from nonprimary infection.
To test the validity of our diagnostic strategy, we
compared the results of congenital infections, presence of
symptoms at birth and appearance of delayed sequelae
obtained from primarily infected mothers previously
identified as having low avidity antibody combined with
positive IgM profile versus the results obtained from
mothers who seroconverted.
The percentage incidence of congenital infections was
very similar in both groups: of 83 women with serocon-
version, 30 (36.1%) had a congenitally infected fetus/
newborn, and, of 286 women with low IgG avidity/
positive IgM, 87 (30.4%) fetuses/newborns were
infected. Very similar results were obtained for the pres-
ence of symptoms (10.8% vs 14.0%) and the appearance
of sequelae (7.2% vs 8.4%).
Because CMV can be detected in the blood of preg-
nant women only during ongoing or recent primary
infections, we tested by antigenemia, PCR, and qPCR
blood samples obtained from 94 pregnant primarily in-
fected women.
Diagnosis of primary CMV infection was made ac-
cording to CMV-specific IgG seroconversion in 23 cases,
and to a combination of IgG low avidity and confirmed
presence of anti-CMV IgM by immunoblot in 71 cases.
Antigenemia was detected in 6 (6.3%), PCR in 22
(33.3.%), and qPCR in 16 (23.8%).
As indicated in Table 2, we calculated the sensitivity,
specificity, positive predictive value (PPV), and negative
predictive value (NPV) of tests in relation to CMV
congenital infection. The statistical analysis of the data
showed that the results obtained with antigenemia, PCR,
and qPCR of maternal blood in relation to congenital
CMV infection gave a poor percentages of sensitivity and
PPV.
Among the 35 pregnant women who transmitted
CMV to the fetus/newborn, only 5 (14.3%) had a posi-
tive determination of pp65 antigenemia at the time of
diagnosis of primary infection with serologic methods.
The antigenemia test had a 98.3% specificity with a
83.3% PPV and 65.9% NPV.
The detection and quantitation of the viral DNA in
maternal blood gave a very similar performance. Only 10
(47.6%) blood samples obtained from 21 pregnant
women who had an infected fetus/newborn were PCR-
positive. This virologic procedure reached the best value
of sensitivity but the worst values of specificity and
PPV—75% and 45.5%—respectively.
In fact, among the 22 positive blood samples for PCR,
12 were collected from mothers who did not transmit
CMV to their newborns and 10 were collected from
mothers who did. In particular, of these 12 specimens, 7
had a low-medium CMV-DNA load (median 350 GE/2
⫻ 105 PMNL; range 120 –1310 GE/2 ⫻ 105 PMNL), 3
specimens values less than 100 GE/2 ⫻ 105PMNL and 2
were not determined.
Of the other 10 PCR-positive specimens collected
from the mothers who transmitted the CMV infection to
their newborns, 9 had a low-medium DNA load (median
586 GE/2 ⫻ 105 PMNL; range 100 –1400 GE/2 ⫻ 105
PMNL) and 1 was not determined.
In the two groups, the median DNA level was very
similar, and the difference was not statistically signifi-
cant (P ⫽ 0.78).
Congenital Cytomegalovirus Infection
TABLE 2 CMV detection in the blood of primarily infected women in relation to congenital infection and
asymptomatic/symptomatic congenital infection
CMV congenital
infection
Yes
Pos 5 1
Antigenemia
pp65-pos cells/2 ⫻ 105
PMNL
Neg 30
Pos 10
PCR-PMNL
Neg 11
ⱖ100 9 7
qPCR
GE/1 ⫻ 105 PMNL
⬍100 18
Abbreviations: CMV ⫽ cytomegalovirus; pos ⫽ positive; neg ⫽ negative; SENS ⫽ sensitivity; SPEC ⫽ specificity; PPV and NP ⫽ positive and negative
predictive value.
DISCUSSION
Diagnosis of primary CMV infection is relatively
straightforward if seroconversion is detected. However,
most pregnant women do not know their prepregnancy
serologic status. How can pregnant women undergoing
an acute CMV infection be identified if this is asymp-
tomatic?
The solution to these problems requires a diagnostic
strategy, employing an algorithm based on screening
pregnant women with a CMV IgG assay and a sensitive
CMV IgM assay, followed by reflex testing of CMV IgM
positive specimens with a CMV IgG avidity assay.
Detection of CMV-specific IgM has traditionally been
considered the most appropriate procedure for screening
pregnant women. However, the search for CMV-specific
IgM has always been hampered by the fact that the
correlation of results obtained with different commercial
kits is poor and results are contradictory if a serum is
tested with different kits [14]. Recently available new
tests may rapidly change this situation because they seem
sensitive and specific enough to detect all pregnant
women undergoing active infection (sensitivity 96%;
specificity 97%) [15].
When anti-CMV IgM is detected, the diagnostic
problems are not over because only 7.5% of IgM-positive
women have a congenitally infected fetus/newborn (Laz-
zarotto T. et al., unpublished data). Moreover, recent
evidence indicates that although titers are lower, IgM
antibody is produced during the acute and late phases of
a primary CMV infection and once produced may persist
for 6 to 9 months. IgM antibody often is produced
during CMV reactivation/reinfection [14, 15]. A positive
413
No SENS SPEC PPV NPV
14.3 98.3 83.3 65.9
58
12
47.6 75 45.5 76.6
36
33.3 82.5 56.3 64.7
33
IgM test should therefore be considered only as the
starting point for a more thorough diagnostic evaluation
to determine whether there is a risk of fetal infection.
Among the procedures differentiating primary from
recurrent infections, the determination of antibody avid-
ity has gained increasing importance [4 –7]. The IgG
avidity test may be of great help in clarifying the clinical
significance of IgM antibody. IgG avidity is indicative of
the low functional affinity of the IgG class antibody.
Early after primary infection, antibodies show a low
avidity for the antigen, but progressively mature acquir-
ing higher avidity. This characteristic is used at diag-
nostic level to discriminate recent primary infections in
several viral diseases. Low IgG avidity is a marker of
primary infection for 18 –20 weeks after onset of symp-
toms in immunocompetent subjects [4, 5].
Recently, we showed that determination of anti-CMV
avidity at 6 –18 weeks’ gestation can identify all women
who would have an infected fetus/newborn (100% sen-
sitivity), whereas at 20 –23 week’s gestation the sensi-
tivity of the test was lower (62.5%) [7].
The percentage incidence of CMV congenital infec-
tions detected in primarily infected pregnant women
identified as having low IgG avidity combined with
positive IgM profile was 30%. This rate of transmission
was very similar (36%) to that of women who serocon-
verted during pregnancy.
The first step of the diagnostic protocol adopted in
this study can be used to gate the decision for prenatal
diagnosis. However, before prenatal diagnostic proce-
dures, the diagnosis of primary infection may be further
414
confirmed or substantially supported by assays for the
detection of virus products in maternal blood.
Data from the literature indicate that the detection of
viral DNA led to diagnosis of primary CMV infection in
all subjects within 1 month after onset, remaining pos-
itive for 6 months, but only about 50% and 25% of the
patients were positive 3– 6 months after infection. Like-
wise, antigenemia and viremia showed a lower sensitiv-
ity, detecting about 50% and 25% of positive patients,
respectively, in the first month after infection [16]. Our
results showed that among the 21 primarily infected
pregnant women who transmitted CMV to the fetus/
newborn, only 10 had a positive detection of viral DNA
in maternal blood at the time of diagnosis of CMV
primary infection by advanced serologic methods.
Moreover, in pregnant women with primary infection
the presence of the virus in the blood did not increase the
risk of intrauterine transmission of CMV infection and
the amount of virus in maternal blood determined by
antigenemia, PCR, and qPCR did not correlate with the
severity of CMV infection in the fetus/newborn (data not
shown). Hence maternal virologic tests are not sufficient
to diagnose a recent primary maternal CMV infection
[16].
Twin pregnancies represent an interesting model to
confirm that the presence of CMV in the blood does not
increase the risk of intrauterine CMV transmission be-
cause different fetuses are exposed to the same maternal
influences at the same time.
We recently described three cases of CMV-infected
twin pregnancies complicated by transmission of the
infection. Two pregnant women with twin pregnancies
and one woman with a triple pregnancy with primary
CMV infection defined by the presence of IgM and low
IgG avidity or by the presence of clinical symptoms and
abnormal liver enzyme values were evaluated. CMV in-
fection was found in six fetuses/newborns, three of whom
were symptomatic [11].
In particular in the first twin pregnancy (a female and
male fetus) with diamniotic-dichorionic separate placen-
tas, maternal DNAemia was CMV-negative. Only the
female fetus was severely infected. In the third twin
pregnancy (two female fetuses) with fused dichorionic
placentas, the maternal blood was antigenemia and
DNAemia positive and CMV infection in only one fetus
was documented [11].
More recently, a new virologic parameter, detection in
blood of immediate-early mRNA by nucleic-acid se-
quence based amplification (NASBA) technology, has
been used for the diagnosis of primary CMV infection in
pregnant women. Immediate-early mRNA appears to be
slightly more sensitive than DNA detection in diagnos-
ing the early phase of primary CMV infection, but
slightly less sensitive in detecting the late phase. It is
T. Lazzarotto et al.
premature to draw any conclusions, but further investi-
gations in this field are warranted [17].
ACKNOWLEDGMENTS
This study was supported in part by grants from the Ministry
of Public Health (Istituto Superiore di Sanità, AIDS Program);
the Ministry of University, Scientific and Technological Re-
search; the St. Orsola Malpighi General Hospital; and the
University of Bologna. We thank Cristiana Grandi and Gab-
riele Lionetti for excellent technical assistance. Anne Collins
edited the English test.
REFERENCES
1. Alford CA, Stagno S, Pass RF, Britt WJ: Congenital and
perinatal cytomegalovirus infections. Rev Infect Dis
12(Suppl 7):S745, 1990.
2. Stagno S, Pass RF, Cloud G, Britt WJ, Henderson RE,
Walton PD, Veren DA, Page F, Alford CA: Primary
cytomegalovirus infection in pregnancy. Incidence, trans-
mission to fetus, and clinical outcome. JAMA 256:1904,
1986.
3. Fowler KB, Stagno S, Pass RF, Britt WJ, Boll TJ, Alford
CA: The outcome of congenital cytomegalovirus in rela-
tion to maternal antibody status. N Engl J Med 326:663,
1992.
4. Grangeot-Keros L, Mayaux MJ, Lebon P, Freymuth F,
Eugene G, Stricker R, Dussaux E: Value of cytomegalo-
virus (CMV) IgG avidity index for the diagnosis of pri-
mary CMV infection in pregnant women. J Infect Dis
175:944, 1997.
5. Lazzarotto T, Spezzacatena P, Pradelli P, Abate DA, Va-
rani S, Landini MP: Avidity of immunoglobulin G direct
against human cytomegalovirus during primary and sec-
ondary infections in immunocompetent and immunocom-
promised subjects. Clin Diagn Lab Immunol 4:469, 1997.
6. Egger M, Bäder U, Enders G: Combination of microneu-
tralization and avidity assays: improved diagnosis of re-
cent primary human cytomegalovirus infection in single
serum sample of second trimester pregnancy. J Med Virol
60:324, 2000.
7. Lazzarotto T, Varani S, Spezzacatena P, Gabrielli L,
Pradelli P, Guerra B, Landini MP: Maternal IgG avidity
and IgM detected by blot as diagnostic tools to identify
pregnant women at risk of transmitting cytomegalovirus.
Viral Immnunol 13:137, 2000.
8. Nielsen SL, Sørensen I, Andersen HK: Kinetics os specific
immunoglobulins M, E, A, and G in congenital, primary,
secondary cytomegalovirus infection studied by antibody-
capture enzyme-linked immunosorbent assay. J Clin Mi-
crobiol 26:654, 1988.
9. Lazzarotto T, Ripalti A, Bergamini G, Battista MC, Spe-
zzacatena P, Campanini F, Pradelli P, Varani S, Gabrielli
L, Maine GT, Landini MP: Development of a new cyto-
megalovirus (CMV) immunoglobulin M (IgM) immuno-
Congenital Cytomegalovirus Infection
blot for detection of CMV-specific IgM. J Clin Microbiol
11:3337, 1998.
10. Vivarelli M, De Ruvo N, Lazzarotto T, Bellusci R, Lan-
dini MP, Varani S, Cavallari A: Abstention from treat-
ment of low-level pp65 cytomegalovirus antigenemia af-
ter liver transplantation: a prospective study.
Transplantation 70:1183, 2000.
11. Lazzarotto T, Gabrielli L, Foschini MP, Lanari M, Guerra
B, Eusebi V, Landini MP: Congenital cytomegalovirus
infection in twin pregnancies: viral load in the amniotic
fluid and pregnancy outcome. Pediatrics 112:E153–7,
2003.
12. Lamy ME, Mulongo KN, Gadisseux JF, Lyon G, Gaudy
V, Van Lierde M: Prenatal diagnosis of fetal cytomegalo-
virus infection. Am J Obstet Gynecol 166:91, 1992.
13. Rothman KJ, Greenland S: Modern Epidemiology. Phil-
adelphia: Lippincott-Raven Publishers, 1998.
14. Genser B, Truschnig-Wilders M, Stunzner D, Landini
415
MP, Halwachs-Baumann G: Evaluation of five commer-
cial enzyme immunoassays for the detection of human
cytomegalovirus-specific IgM antibodies in the absence of
a commercially available gold standard. Clin Chem Lab
Med 39:62, 2001.
15. Maine GT, Lazzarotto T, Landini MP: New developments
in the diagnosis of maternal and congenital CMV infec-
tion. Expert Rev Mol Diagn 1:19, 2001.
16. Revello MG, Zavattoni M, Sarasini A, Percivalle E, Simo-
ncini L, Gerna G: Human cytomegalovirus in blood of
immunocompetent persons during primary infection:
prognostic implications for pregnancy. J Infect Dis 177:
1170, 1998.
17. Revello MG, Lilleri D, Zavattoni M, Stronati M, Bollani
L, Middeldorp JM, Gerna G: Human cytomegalovirus
immediate-early messenger RNA in blood of pregnant
women with primary infection and of congenitally in-
fected newborns. J Infect Dis 184:1078, 2001.