Journal of Clinical Virology 64 (2015) 45–51

Contents lists available at ScienceDirect

Journal of Clinical Virology

journal homepage: www.elsevier.com/locate/jcv

Primary human cytomegalovirus infections: Kinetics of ELISA-IgG

and neutralizing antibody in pauci/asymptomatic pregnant
women vs symptomatic non-pregnant subjects
Chiara Fornara a , Milena Furione b , Daniele Lilleri a , Ilaria Cane a , M. Grazia Revello c ,
Maurizio Zavattoni b , Giuseppe Gerna a,∗
a
Laboratori Sperimentali di Ricerca, Area Trapiantologica, Fondazione Istituto di Ricovero e Cura a Carattere Scientifico Policlinico San Matteo,
27100 Pavia, Italy
b
Struttura Semplice Virologia Molecolare, Struttura Complessa Microbiologia e Virologia, Fondazione Istituto di Ricovero e Cura a Carattere
Scientifico Policlinico San Matteo, 27100 Pavia, Italy
c
Struttura Complessa Ostetricia e Ginecologia, Università degli Studi di Pavia, Istituto di Ricovero e Cura a Carattere Scientifico Policlinico
San Matteo, 27100 Pavia, Italy

a r t i c l e i n f o a b s t r a c t

Article history: Background: Human cytomegalovirus infections are mostly asymptomatic in infants and young children,
Received 10 June 2014 while they are often associated with overt clinical symptoms in adults.
Received in revised form Objectives: To verify whether the antibody response to HCMV is more potent in symptomatic non-
17 December 2014
pregnant adults as compared to asymptomatic/paucisymptomatic pregnant women.
Accepted 3 January 2015
Study design: Overall, 36 consecutive pregnant women with primary HCMV infection were compared
with 10 consecutive symptomatic non-pregnant subjects with primary HCMV infection and overt
Keywords:
clinical symptoms. Levels of IgG antibody responses to HCMV-infected cell lysate and the pentamer
Symptomatic/asymptomatic HCMV
infections
gH/gL/pUL128L, gH/gL and gB HCMV glycoprotein complexes as well as neutralizing antibodies pre-
HCMV glycoprotein complexes venting infection of epithelial cells (ARPE-19) and human embryonic lung fibroblast (HELF) cells were
Neutralizing antibodies compared at intervals of 1–30, 31–60, 61–90, 91–180 and 181–360 days after onset of infection. In parallel,
ELISA-IgG antibodies viral load was quantified by real-time PCR.
HCMV load Results: In symptomatic non-pregnant subjects, the IgG responses to HCMV lysate as well as to gH/gL and
ARPE-19 neutralizing antibodies were significantly higher from 31 to 60 through 180 days after infection
onset. In the same patients, the IgG antibody responses to the pentamer and HELF-neutralizing antibody
were significantly higher starting 90 days post-infection.
Conclusions: The presence of overt clinical symptoms is associated with a significantly higher antibody
response (concomitantly with a higher viral load) in non-pregnant subjects with symptomatic primary
HCMV infection as compared to pregnant women with paucisymptomatic/ asymptomatic primary infec-
tion (and lower viral load).
© 2015 Elsevier B.V. All rights reserved.

1. Background

Primary human cytomegalovirus (HCMV) infection mostly

Abbreviations: HCMV, human cytomegalovirus; HCMV IgG, IgG antibody to
HCMV; gH/gL/pUL128L-IgG, ELISA IgG to the pentamer; gH/gL-IgG, ELISA IgG to the
dimer gH/gL; gB-IgG, ELISA IgG to HCMV gB; ARPE-19-Nt Ab, neutralizing antibody
preventing infection of ARPE-19 epithelial cells; HELF-Nt Ab, neutralizing antibody
preventing infection of human embryonic lung fibroblasts.
∗ Corresponding author at: Laboratori Sperimentali di Ricerca, Area Trapianto-
logica, Fondazione IRCCS Policlinico, Area Trapiantologica, Fondazione Istituto di
Ricovero e Cura a Carattere Scientifico Policlinico San Matteo, 27100 Pavia, Italy.
Tel.: +39 0382 503314; fax: +39 0382 502648.
E-mail address: g.gerna@smatteo.pv.it (G. Gerna).
occurs in an asymptomatic presentation in infants and young chil-
dren, while it is more often associated with overt clinical symptoms
in adults. The risk of virus transmission to the fetus during primary
HCMV infection, ranging around 40%, is well recognized [1,2]. As a
consequence, even minor or mild clinical symptoms often draw the
attention of well-informed pregnant women and medical doctors
in an effort to identify the potential cause of such symptoms. On
the contrary, outside of pregnancy, mild symptoms may be largely

http://dx.doi.org/10.1016/j.jcv.2015.01.004
1386-6532/© 2015 Elsevier B.V. All rights reserved.
46 C. Fornara et al. / Journal of Clinical Virology 64 (2015) 45–51
overlooked, limiting requests for medical intervention or hospital
admission to cases with overt clinical symptoms.
2. Objectives
To study the relationship of clinical symptom severity and the
level of antibody response in a group of consecutive primary HCMV
infections associated with overt clinical symptoms in non-pregnant
subjects vs a group of consecutive primary HCMV infections asso-
ciated with no or mild clinical symptoms in pregnant women
(asymptomatic/paucisymptomatic infections). In other words, the
main goal of this study was to verify whether a correlation exists
between the severity of clinical symptoms (and viral replication)
and the level of the antibody response (measured with different
antibody assays).
3. Study design
3.1. Patient enrollment
The study design consisted of the analysis of the antibody
response to HCMV infection during 2011–2013 in (i) 36 consecutive
pregnant women with primary HCMV infection either associated
with mild clinical symptoms (n = 26) and moderate (<38 ◦ C) fever
lasting <2 weeks (n = 10) or no fever (n = 16), while 10 subjects
were totally asymptomatic. All subjects were followed for one year,
monthly until 6 months, then at 9 and 12 months after infection
onset. A preliminary analysis showed that there was no a significant
difference in the antibody response between asymptomatic and
paucisymptomatic pregnant women; and (ii) 10 age-matched non-
pregnant subjects (7 males, 3 females) with a fully symptomatic
(fever >38 ◦ C lasting >2 weeks) primary infection.
3.2. Diagnosis of primary infection
Primary infection was diagnosed based on at least two of the
following criteria: HCMV-specific IgG seroconversion, presence
of virus-specific IgM antibody, low IgG avidity index (AI), and
DNAemia. When a pregnant woman with a suspected primary
HCMV infection was referred to our institution, HCMV-specific IgG
and IgM antibody were routinely determined by ETI-CYTOK-G and
ETI-CYTOK-M (DiaSorin, Saluggia, Italy), respectively (from here on,
these assays will be named conventional assays). IgG avidity was
determined by an in-house developed ELISA assay [3]. Infection
onset timing was based on detection of seroconversion and other
serologic/virologic findings, whenever possible in association with
presence of clinical symptoms [4,5].
3.3. Viral load determination
Viral load quantification during follow-up was obtained by
real-time PCR (detection limit 25 copies/ml whole blood). The con-
version factor of copies into IU was possible with reference to the
WHO International HCMV DNA Standard [6].
3.4. IgG antibody determination
Glycoprotein complexes gH/gL/pUL128L (pentamer), gH/gL, and
gB were prepared and purified as reported [7]. Different con-
structs were used to transfect HEK293F cells (Invitrogen). After
10-days culture, supernatants were harvested and stored at –20 ◦ C.
Microplates coated with an in-house developed murine anti-gH
mAb (H1P73), or an anti-gB mAb (HCMV 37, Abcam, Cambridge,
UK) were incubated with cell culture supernatants containing the
relevant glycoprotein complex. After addition of human serum,
the peroxidase-labeled goat IgG fraction to human IgG (Fc-chain-
specific) was added prior to addition of the substrate solution.
A cut-off of 0.10 (OD) for both the pentamer-IgG and gH/gL-IgG
assays, and a cut-off of 0.20 for the gB-IgG assay were calculated
(from here on, these assays will be named novel assays).
3.5. Neutralization assays
Virus (VR1814)-human serum antibody mixtures were incu-
bated onto monolayers of ARPE-19 epithelial cells or human
embryonic lung fibroblast (HELF) cells, and centrifuged at 700 × g
for 30 min, as reported [8]. After 48 h incubation, cells were fixed
and stained using a pool of p72-specific murine mAbs [9]. The
serum dilution inhibiting virus infectivity by 50% or more with
respect to the virus control was considered the 50% neutraliz-
ing antibody titer. The ARPE-19 neutralization (ARPE-19-Nt) assay
will be included in this study into the group of novel (unconven-
tional) assays, while the HELF neutralization (HELF-Nt) assay will
be included into the group of conventional assays.
3.6. Statistical analysis
IgG and IgM antibody titers to HCMV lysate, IgG AI, as well as IgG
antibody titers to the pentamer gH/gL/pUL128L, gH/gL, and gB, and
neutralizing antibody titers preventing infection of both epithelial
and fibroblast cells were compared by the Mann–Whitney U-test
for each time interval. Non-linear regression models were used to
express the kinetics of the same serological parameters. The curves
were compared by the extra-sum of square F test (Graph Pad Prism
5.0 software).
4. Results
4.1. Patients’ characteristics (demographic, clinical and
diagnostic) of the populations studied
No significant difference was found for age and week of ges-
tation, as well as for time to diagnosis between non-pregnant vs
pregnant subjects. However, the most consistent difference was
relevant to clinical symptoms (Table 1).
4.2. Incidence of clinical symptoms/signs in the two populations
examined
In the 10 symptomatic non-pregnant subjects, high fever
(>38 ◦ C) lasting 2–4 weeks was consistently detected, in associa-
tion with asthenia in 6 (60%) patients and headache in additional
6 (60%). On the other hand, in the 26 paucisymptomatic pregnant
women fever was moderate (≤38 ◦ C) and short-lasting (<2 weeks)
in 10 (38%) subjects, whereas the other 16 women had no fever,
but some other mild symptoms. Other recurrent symptoms were
asthenia in 9 (35%), headache in 9 (35%) and pharyngitis in 7 (27%)
women. In both groups, other symptoms/signs were sporadically
observed, such as lymphoadenopathy, cough, rhinitis, arthromyal-
gias, lymphocytosis, elevated liver enzyme levels. Thus, the most
differentiating symptom between the two groups was a high long-
lasting fever in the symptomatic non-pregnant group. Finally, 10
pregnant women were totally asymptomatic (Table 1).
4.3. Differential kinetics of the HCMV-IgG, gH/gL-IgG and
ARPE-19 neutralizing antibody responses in symptomatic
non-pregnant individuals vs pauci/asymptomatic pregnant
women
Starting 30 days after onset of infection, a significantly higher
antibody titer was found in non-pregnant subjects until 180–360
 C. Fornara et al. / Journal of Clinical Virology 64 (2015) 45–51 47

Table 1
Demographic and clinical characteristics of pregnant and non-pregnant subjects examined in the study.

Patients Age Sex Week of gestation at Symptoms Days between infection

(median, infection onset onset and first lab test
range) (median, range) (i.e., study entry)

Pregnant
Mildly symptomatic (n = 26) 34 (29–42) F 16 (1–25) Moderate fever 18 (6–36)
(≤38 ◦ C) lasting <2
weeks (38%), asthenia
(35%), headache (35%),
pharyngitis (27%)
Asymptomatic (n = 10) 33.5 (24–40) F 19 (8–26) NAa 19 (3–43)

Non-pregnant
Symptomatic (n = 10) 37.5 (23–74) 7 M, 3 F NAa High fever (>38◦ ) 11.5 (3–60)
lasting 2–4 weeks, in
association with
asthenia (60%), and
headache (60%)
a
NA – not applicable.

days p.i. according to the different antibody assays: IgG to HCMV
lysate (Fig. 1), IgG to gH/gL (Fig. 2), and ARPE-19-Nt antibodies
(Fig. 3). The significantly higher antibody titer in symptomatic non-
pregnant individuals interested the entire time interval of 91–360
days for IgG to HCMV and ARPE-19 Nt, and the 91–180 day period
for IgG to gH/gL. These findings indicate a persistent difference
between the two groups of patients for all three types of antibody
response. No significant changes in the antibody response between
the 2 groups of pregnant and non-pregnant subjects was observed
in the time interval of 181–360 days either including in the group
of pregnant women only sera collected during pregnancy or after
delivery (Figs. 1–3).

Fig. 1. (A) kinetics of the IgG antibody titer to HCMV (expressed in international
units) in the two patient populations studied. (B) Medians of the IgG antibody levels
in pregnant and non-pregnant subjects at different intervals after onset of infection.
Statistical difference is reported at the top of each interval when significant (from
day 31 to day 360 p.i.).
Fig. 2. (A) Kinetics of the ELISA IgG antibody titer to gH/gL after onset of infection in
pregnant and non-pregnant subjects. (B) Medians of IgG antibody levels at different
intervals after onset of infection. Significantly statistical differences are reported at
the top of each interval (from day 31 to day 180 p.i.).
48 C. Fornara et al. / Journal of Clinical Virology 64 (2015) 45–51
Fig. 3. (A) Kinetics of the VR1814 neutralizing antibody titer in ARPE-19 epithelial
cells in the two groups of subjects examined. (B) Medians of the ARPE-19 neu-
tralizing antibody titer at different intervals after onset of infection. Statistically
significant differences are reported at the top of each interval (from day 31 to day
360 p.i.).
In addition, no difference (within the limits of this series) was
observed between the two groups for IgM antibody to HCMV and
AI (among conventional assays) and for IgG antibody to gB (among
novel assays).
4.4. Differential kinetics of IgG to the pentamer and
HELF-neutralizing antibody responses in the two groups of
patients with primary HCMV infection
Unlike the results reported above, a significantly different
(in magnitude) antibody response between symptomatic non-
pregnant subjects and pregnant women was observed more than
90 days after infection onset when IgG antibodies to the pentamer
(Fig. 4), and HELF-Nt antibodies (Fig. 5) were measured. This differ-
ence was observed in the interval of 91–360 days for IgG antibodies
to the pentamer and 91–180 days for HELF-Nt antibodies. In addi-
tion, no significant changes in the antibody response between the
2 groups of pregnant and non-pregnant subjects was observed in
the time interval of 181–360 either including in the group of preg-
nant women only sera collected during pregnancy or after delivery
(Figs. 4–5).
Fig. 4. (A) Kinetics of the ELISA IgG antibody titer to the pentamer gH/gL/pUL128L in
pregnant and non-pregnant subjects. (B) Medians of the ELISA IgG antibody response
to the pentamer at different intervals after onset of infection. Statistically significant
differences are reported at the top of each interval starting 91 days p.i.
4.5. Viral load
The differences observed in antibody titers between symp-
tomatic non-pregnant individuals and pregnant women suggested
to compare the viral load levels observed in the two groups of
patients (within the limits of the variable intervals of time between
sequential samples) as a potential indicator of differential viral
replication (Fig. 6). Results showed that a significant (p < 0.01) dif-
ference in viral load does exist between symptomatic non-pregnant
patients (median 188, range 25–4,200, DNA copies/ml blood) and
paucisymptomatic/asymptomatic pregnant women (median 25,
range 25–6,762, copies). Thus, concomitantly with the higher
antibody titer, a higher viral load was detected in non-pregnant
individuals.
4.6. Sequential antibody responses in individual patients
From the individual cases of primary HCMV infection, in which
early blood samples were available, it was possible to confirm
[7] the sequential pattern of early antibody responses accord-
ing to the novel assays (data not reported). The first antibodies
detected were IgG to gB, followed by ARPE-19-Nt antibodies and
IgG antibodies to the pentamer, then IgG antibodies to gH/gL,
and finally HELF-Nt antibodies. Conventional IgG and IgM anti-
 C. Fornara et al. / Journal of Clinical Virology 64 (2015) 45–51 49

Table 2
Kinetics of the different antibody responses to primary HCMV infection by different assays in a symptomatic non-pregnant subject and an asymptomatic pregnant woman.

Patient, age Days after onset DNAa IgGb IgMc AId Pentamer IgG gH/gL IgG gB IgG ARPE-19 Nte HELF Nt

# 6, 29 years 15 100 0.7 11.4 15 363 <50 1113 1280 <5

Symptomatic 25 25 ND ND ND 1489 99 15021 2560 ND
Non- 61 25 3.2 2.8 38 6300 1010 32699 5120 40
pregnant 97 Neg 4.3 <0.9 58 8142 2621 34988 10240 160
201 Neg 4.4 <0.9 78 21510 4225 41452 20480 640
346 ND ND ND ND 37760 8877 53419 40960 160

# 3, 32 years 28 25 1.3 2.9 28 142 <50 9718 1280 <5

Asymtomatic 54 25 1.5 1.5 36 4562 1705 15799 1280 20
Pregnant 125 Neg 2.4 <0.9 65 10219 8750 38317 5120 160
women 190 Neg 2.6 <0.9 75 17065 6494 21641 5120 640
385 ND ND ND ND 24985 7443 25369 10240 640
a
Copies/ml blood.
b
IgG, pos >0.4 IU/ml.
c
IgM, pos >1.1 (index).
d
AI, avidity index (%).
e
Nt, neutralization test.

bodies to whole virus were among the earliest antibodies to 5. Discussion

appear.
In addition, as an example, one case of symptomatic and one
of asymptomatic infection are reported in Table 2, with the actual
titers obtained by conventional and novel assays.
Following the first detection of HCMV in the 50 s, HCMV infec-
tion in the immunocompetent host was considered clinically silent
in approximately 90% of acquired primary infections, although
it was known that the virus could cause an acute febrile ill-
ness resembling Epstein–Barr virus mononucleosis [10]. At that
time, several groups reported a HCMV mononucleosis prevalence
ranging around 5% of cases of primary infection [11–13]. The
most predominant clinical symptoms reported [10,14,15] were:
fever lasting 2 weeks (95%) or 3 weeks (∼50% of cases) with
malaise, fatigue and myalgia, followed by cervical lymphoadenopa-
thy (6–56%), sore throat (56%), hepatosplenomegaly (0–53%) and
rash (0–44%). Predominant signs included elevated liver enzymes
(78–100%) and lymphocytosis (>5,000/␮l blood) in 78–95% of cases.
In the following years, a more accurate case history investigation
conducted on large series of pregnant women allowed the detection
of mild or recurrent clinical symptoms and signs in the majority
of pregnant women with primary infection. In three subsequent
surveys conducted by our group, the rate of clinical symptoms/signs
was found to increase from 38/60 (60%) [16] to 166/244 (68%) [4],
and finally to 530/721 (73%) [5]. This increasing incidence was likely
due to the progressively more accurate clinical investigation.

Fig. 5. (A) Kinetics of the AD169 neutralizing antibody titer in HELF in the two groups Fig. 6. Median DNA levels in pregnant and non-pregnant subjects. The highest DNA
of patients. (B) Medians of the HELF neutralizing antibody titer to AD169 at different level detected was selected from 36 pregnant women and 10 non-pregnant subjects.
intervals after onset of infection. Statistically significant difference is reported at the The difference, as indicated, is statistically significant. Detection limit of 25 copies/ml
top of the 91–180 day interval. is indicated. Undetect., undetectable.
50 C. Fornara et al. / Journal of Clinical Virology 64 (2015) 45–51
In the present study, we had the opportunity to observe in
symptomatic non-pregnant patients with primary HCMV infec-
tion the following major distinctive features with respect to
pauci/asymptomatic pregnant women: (i) clinical symptoms were
more severe, in particular with respect to fever which was higher
(>38 ◦ C) and longer lasting; (ii) viral load observed was significantly
higher; (iii) both the ELISA-IgG to viral glycoprotein complexes
and neutralizing antibody titers were significantly higher in
non-pregnant patients compared to pregnant women either prior
to or after delivery, starting 30, or 90 days after onset of infection,
according to different assays.
The role of antibodies in protection from HCMV reactivation
(in association with T-cell immunity) remains substantially to be
defined. However, it has been reported in uncontrolled studies
that HCMV hyperimmune globulin may prevent virus transmis-
sion to the fetus in seronegative pregnant women, and may also
improve symptoms of congenitally infected fetuses [17]. Although
the conclusions of this study have not been confirmed by a recent
randomized, double-blind controlled study [2], antibody pres-
ence still seems to be associated with a much lower risk of virus
transmission to fetus [4] with special reference to some types of
antibodies limiting virus spread [7]. Among antibodies potentially
protecting from HCMV infection, those endowed with neutraliz-
ing activity directed to the pentamer have been shown to be much
more potent than those directed to gH and gB [18].
In more detail, ELISA IgG antibodies to whole virus and ELISA IgG
antibodies to gH/gL were detected at significantly higher titers in
non-pregnant subjects between 30 and 180 days, and ARPE-19 Nt
antibodies between 30 and 360 days. Thus, IgG antibodies to whole
virus and to gH/gL were among the earliest and most efficient anti-
bodies in differentiating the two patient populations examined,
as well as ARPE-19-Nt antibodies expressing their differentiating
potential from 30 days through 360 days after onset of infection.
A delayed discriminating capacity was shown by IgG antibodies to
the pentamer (91 through 360 days) and HELF-Nt antibodies (91
through 180 days). In this respect, it is somewhat surprising that
gB antibodies did not differentiate the two populations studied at
any time interval examined, as well as IgM antibodies and AI. As
previously defined and confirmed (7) by this study, in HCMV pri-
mary infection, the antibodies detected by different assays appear
according to a defined sequential temporal pattern: first to appear
are IgG antibodies to gB, and HELF-Nt antibodies are the last to
appear. As a result, no close correlation appears to exist between
early antibody detection and early discriminating capacity of pri-
mary HCMV infections of different clinical severity. On the contrary,
an increase in antibody titer observed in symptomatic infections
appears to be a phenomenon that develops at later times (1–3
months) after infection onset.
The rationale behind the differential antibody response likely
may reside in the higher virus replication rate in symptomatic
non-pregnant patients. In a previous study, a correlation of HCMV
excretion and IgG levels was reported [19]. However, immuno-
competent patients with primary HCMV infection could not be
monitored starting from the onset of infection and at short time
intervals (as it can be done with transplant recipients), thus it
is likely that we missed the true peak of viral DNA. The results
reported in this study refer to viral load levels measured by chance
during the clinical follow-up. Other potential factors, such as preg-
nancy and time to diagnosis might be considered to explain the
different antibody response. However, this study does not seem to
support the significant role of these additional factors. As for preg-
nancy we compared, for the time interval of 181–360 days, levels
of total HCMV-specific IgG, neutralizing antibodies on epithelial
cells and IgG to the pentamer, either obtained during pregnancy or
after delivery with levels obtained in non-pregnant subjects. Total
IgG levels after delivery were not significantly different from those
observed before delivery, and were significantly lower than in non-
pregnant subjects. Levels of neutralizing antibodies on epithelial
cells and IgG antibodies to the pentamer were not significantly dif-
ferent among the three groups. However, the fact that antibody
levels measured after delivery were in the same range as those
measured before delivery seems to rule out a major role of preg-
nancy in suppressing the antibody response [20]. In addition, the
different time to diagnosis (shorter, although not significantly, in
symptomatic subjects than in pregnant women, see Table 1) does
not appear to influence the antibody response.
The clinical epidemiological basis for such a different behav-
ior is likely due to the fact that the majority of pregnant women
are either diagnosed with primary HCMV infection based on rou-
tine laboratory testing when asymptomatic, or they pay continuous
focused attention to each minimal symptom emerging during preg-
nancy, thus receiving timely medical intervention (both clinical
examination and laboratory testing). Among pregnant women,
marital status, high parity, history of blood transfusion and age
were found to be significant risk factors for HCMV infection (20).
On the other hand, non-pregnant patients are less prone to worry
over minor symptoms and generally request medical interven-
tion only when major symptoms (such as high fever) appear. As a
result, paucisymptomatic/asymptomatic HCMV infections in non-
pregnant patients run their course mostly undiagnosed without
requiring medical intervention. On the contrary, mild symptoms
in pregnant women may attract medical attention and trigger the
adequate diagnostic procedures. Obviously, when symptoms are
totally absent, timing of HCMV infection may become precarious
and uncertain, and the physician must take into consideration the
time-course of the antibody response according to different assays,
as reported in this study.
In conclusion, we observed that symptomatic HCMV infec-
tions in the immunocompetent host are characterized by a higher
viral load than asymptomatic and mildly symptomatic infections
in pregnant women. This finding observed in immunocompetent
individuals resembles what is currently observed in transplant
recipients with primary or recurrent HCMV infection, in whom
symptomatic HCMV disease is associated with a much higher viral
load than asymptomatic infections [21,22]. However, HCMV load
observed in immunocompetent subjects is by far lower than that
observed in the immunocompromised host [23]. As a consequence
of the higher antigen exposure occurring in symptomatic subjects,
the immune system receives greater and longer-lasting stimu-
lation, resulting in the development of higher titer anti-HCMV
antibodies.
Funding
This work was supported by the Fondazione Carlo Denegri,
Torino, Italy (to G. Gerna), Fondazione CARIPLO, Milan, Italy grant
93,043/A (to G. Gerna), and Ricerca Finalizzata grant GR-2010-
2311329 from the Ministero della Salute, Rome, Italy (to D. Lilleri).
Competing interests
None declared.
Ethical approval
This study was approved by the Ethics Committee of the Fon-
dazione IRCCS Policlinico San Matteo (Procedure P-20100035854).
 C. Fornara et al. / Journal of Clinical Virology 64 (2015) 45–51
Acknowledgements
The authors thank Daniela Sartori for careful editing of the
manuscript and Laurene Kelly for revision of the English.
References
[1] K.B. Fowler, S. Stagno, R.F. Pass, W.J. Britt, T.J. Boll, C.A. Alford, The outcome of
congenital cytomegalovirus infection in relation to maternal antibody status,
N. Engl. J. Med. 326 (1992) 663–667.
[2] M.G. Revello, T. Lazzarotto, B. Guerra, A. Spinillo, E. Ferrazzi, A. Kustermann,
et al., A randomized trial of hyperimmune globulin to prevent congenital
cytomegalovirus, N. Engl. J. Med. 370 (2014) 1314–1324.
[3] M.G. Revello, E. Genini, G. Gorini, C. Klersy, A. Piralla, G. Gerna, Comparative
evaluation of eight commercial human cytomegalovirus IgG avidity assays, J.
Clin. Virol. 48 (2010) 255–259.
[4] M.G. Revello, G. Gerna, Diagnosis and management of human cytomegalovirus
infection in the mother, fetus, and newborn infant, Clin. Microbiol. Rev. 15
(2002) 680–715.
[5] M.G. Revello, E. Fabbri, M. Furione, M. Zavattoni, D. Lilleri, B. Tassis, et al., Role
of prenatal diagnosis and counseling in the management of 735 pregnancies
complicated by primary human cytomegalovirus infection: a 20-year
experience, J. Clin. Virol. 50 (2011) 303–307.
[6] M. Furione, V. Rognoni, E. Cabano, F. Baldanti, Kinetics of human
cytomegalovirus (HCMV) DNAemia in transplanted patients expressed in
International Units as determined with the Abbott RealTime CMV assay and
an in-house assay, J. Clin. Virol. 55 (2012) 317–322.
[7] D. Lilleri, A. Kabanova, M.G. Revello, E. Percivalle, A. Sarasini, E. Genini, et al.,
Fetal human cytomegalovirus transmission correlates with delayed maternal
antibodies to gH/gL/pUL128-130-131 complex during primary infection, PLoS
ONE 8 (2013) e59863.
[8] G. Gerna, A. Sarasini, M. Patrone, E. Percivalle, L. Fiorina, G. Campanini, et al.,
Human cytomegalovirus serum neutralizing antibodies block virus infection of
endothelial/epithelial cells, but not fibroblasts, early during primary infection,
J. Gen. Virol. 89 (2008) 853–865.
[9] G. Gerna, F. Baldanti, E. Percivalle, M. Zavattoni, G. Campanini, M.G. Revello,
Early identification of human cytomegalovirus strains by the shell vial assay is
prevented by a novel amino acid substitution in UL123 IE1 gene product, J.
Clin. Microbiol. 41 (2003) 4494–4495.
[10] E. Klemola, R. Von Essen, G. Henle, W. Henle, Infectious-mononucleosis-like
disease with negative heterophil agglutination test: clinical features in
relation to Epstein–Barr virus and cytomegalovirus antibodies, J. Infect. Dis.
121 (1970) 608–614.
51
[11] S. Stagno, R.F. Pass, G. Cloud, W.J. Britt, R.E. Henderson, P.D. Walton, et al.,
Primary cytomegalovirus infection in pregnancy incidence, transmission to
fetus, and clinical outcome, JAMA 256 (1986) 1904–1908.
[12] P.D. Griffiths, C. Baboonian, A prospective study of primary cytomegalovirus
infection during pregnancy: final report, Br. J. Obstet. Gynaecol. 91 (1984)
307–315.
[13] C.S. Peckham, K.S. Chin, J.C. Coleman, K. Henderson, R. Hurley, P.M. Preece,
Cytomegalovirus infection in pregnancy: preliminary findings from a
prospective study, Lancet 1 (1983) 1352–1355.
[14] M.C. Jordan, W. Rousseau, J.A. Stewart, G.R. Noble, T.D. Chin, Spontaneous
cytomegalovirus mononucleosis: clinical and laboratory observations in nine
cases, Ann. Intern. Med. 79 (1973) 153–160.
[15] C.S. Pannuti, L.S. Vilas Boas, M.J. Angelo, V. Amato Neto, G.C. Levi, J.S. de
Mendonca, et al., Cytomegalovirus mononucleosis in children and adults:
differences in clinical presentation, Scand. J. Infect. Dis. 17 (1985)
153–156.
[16] M.G. Revello, G. Gerna, Diagnosis and implications of human cytomegalovirus
infection in pregnancy, Fetal. Matern. Med. Rev. 11 (1999)
117–134.
[17] G. Nigro, S.P. Adler, R. La Torre, A.M. Best, Passive immunization during
pregnancy for congenital cytomegalovirus infection, N. Engl. J. Med. 353 (2005)
1350–1362.
[18] A. Macagno, N.L. Bernasconi, F. Vanzetta, E. Dander, A. Sarasini, M.G. Revello,
et al., Isolation of human monoclonal antibodies that potently neutralize
human cytomegalovirus infection by targeting different epitopes on the
gH/gL/UL128-131A complex, J. Virol. 84 (2010) 1005–1013.
[19] A. Natali, P. Valcavi, M.C. Medici, E. Dieci, S. Montali, C. Chezzi,
Cytomegalovirus infection in an Italian population: antibody prevalence, virus
excretion and maternal transmission, Microbiologica 20 (1997)
123–133.
[20] Z. Maingi, A.K. Nyamache, Seroprevalence of cytomegalovirus (CMV) among
pregnant women in Thika, Kenya, BMC Res. Notes 7 (2014)
794.
[21] P. Grossi, L. Minoli, E. Percivalle, W. Irish, M. Viganò, G. Gerna, Clinical and
virological monitoring of human cytomegalovirus infection in 294 heart
transplant recipients, Transplantation 59 (1995) 847–851.
[22] A.V. Cope, C. Sabin, A. Burroughs, K. Rolles, P.D. Griffiths, V.C. Emery,
Interrelationships among quantity of human cytomegalovirus (HCMV) DNA in
blood, donor-recipient serostatus, and administration of methylprednisolone
as risk factors for HCMV disease following liver transplantation, J. Infect. Dis.
176 (1997) 1484–1490.
[23] D. Lilleri, P. Zelini, C. Fornara, G. Comolli, M.G. Revello, G. Gerna, Human
cytomegalovirus-specific CD4+ and CD8+ T cell responses in primary infection
of the immunocompetent and the immunocompromised host, Clin. Immunol.
131 (2009) 395–403.