Seroprevalence and Risk Factors for Cytomegalovirus Infections
in Adolescent Females
Laura Patricia Stadler,1David I. Bernstein,2S. Todd Callahan,3Christine B. Turley,5Flor M. Munoz,6
Jennifer Ferreira,9Mekhala Acharya,9Gina A. Gorgone Simone,9Shital M. Patel,7,8
Kathryn M. Edwards,4and Susan L. Rosenthal10,11
1Department of Pediatrics, Division of Infectious Diseases, University of Kentucky, Lexington;2Department of
Pediatrics, Division of Infectious Diseases, Cincinnati Children’s Hospital Medical Center, Ohio;3Department of
Adolescent and Young Adult Medicine, Vanderbilt University, and4Department of Pediatrics, Division of Pediatric
Infectious Disease, Vanderbilt Vaccine Research Program, Nashville, Tennessee;5Department of Pediatrics,
University of Texas Medical Branch at Galveston, Departments of6Pediatrics,7Medicine, and8Molecular Virology
and Microbiology, Baylor College of Medicine, Houston, Texas;9EMMES Corporation, Rockville, Maryland;
10Department of Pediatrics, Columbia University Medical Center, and11New York-Presbyterian Hospital Morgan
Stanley Children’s Hospital, New York, New York
Corresponding Author: Laura Patricia Stadler, MEd, MD, MS, Assistant Professor of Pediatrics,
Division of Infectious Disease, KY Clinic Room J414, 740 S Limestone St, Lexington, KY 40536-0284.
Received February 8, 2012; accepted June 28, 2012; electronically published August 3, 2012.
sensorineural hearing loss, developmental delay, and mental retardation. Understanding risk factors for
acquisition of CMV infection in adolescent females will help determine vaccine strategies.
Methods.Females (12–17 years) were recruited from primary care settings in Cincinnati, Galveston,
Houston, and Nashville from June 2006 to July 2010 for a seroepidemiologic study, from which
seronegative participants were recruited for a CMV vaccine trial. Participants (n =1585) responded to
questions regarding potential exposures. For those with young children in the home (n =859),
additional questions were asked about feeding and changing diapers, and for those > 14 years of age (n
=1162), questions regarding sexual activity were asked. Serum was evaluated for CMV antibody using
a commercial immunoglobulin G assay.
Results.Cytomegalovirus antibody was detected in 49% of participants. In the univariate analyses,
CMV seroprevalence was significantly higher among African Americans, those with children <3 years
of age in the home, and those with a history of oral, anal, or vaginal intercourse. Among those with
young children in the home, feeding children and changing diapers further increased the association
with CMV infection. However, in the final multivariate analysis, only African Americans and
household contact with young children were associated with CMV infection.
Conclusions.By age 12, evidence of CMV infection was common. Multiple factors regarding race
and personal behaviors likely contribute to seroconversion earlier in life.
Congenital cytomegalovirus (CMV) is a leading cause of disability, including
Adolescents; Computer-Assisted Screening Interview; CASI; Cytomegalovirus; CMV;
Journal of the Pediatric Infectious Diseases Society, Vol. 2, No. 1, pp. 7–14, 2013. DOI:10.1093/jpids/pis076
© The Author 2012. Published by Oxford University Press on behalf of the Pediatric Infectious Diseases Society.
All rights reserved. For Permissions, please e-mail: email@example.com.
Cytomegalovirus (CMV) is a common infection with
an estimate of seroprevalence in adolescents of 47%–
89% [1–6]. Congenital CMV infection is the leading
cause of birth defects and disability in the United
States , affecting an estimated 8000 infants per year
with sensorineural deafness, mental retardation, and/
or seizure disorders [8–10], and an annual estimated
financial burden of $1–2 billion . The Institute of
Medicine indicated a vaccine targeting 12 year olds
would be highly cost-effective and designated it as a
level I priority .
Because encouraging results from recent trials have
renewed interest in CMV vaccines [13–16], it is impor-
tant to learn more about the risk factors associated
with acquisition of CMV, particularly in young women
12–17 years of age who may have increased rates of in-
fection . Previous studies indicate that CMV may be
transmitted via saliva, urine (or handling of diapers),
blood, cervical secretions, and semen [2, 17–23].
Because young women frequently engage in behaviors
with children at the peak age of CMV shedding (1–2
year olds), they are at risk for coming into contact with
body fluids containing CMV. This suggests that expo-
sure to young children could be a major risk factor for
transmission [24–25]. Therefore, we aimed to evaluate
specific behaviors that may place adolescent females at
increased risk of infection.
MATERIALS AND METHODS
Study Population and Design
To determine the seroprevalence of CMV and factors
associated with CMV infection, females (12–17 years)
were recruited from established patient populations of
adolescent outpatient clinics affiliated with Baylor
College of Medicine ([BCM] Houston, TX), Cincinnati
Cincinnati, OH), The University of Texas Medical
Branch at Galveston, TX ([UTMB] Galveston, TX),
and Vanderbilt University Medical Center ([VUMC]
Nashville, TN) from June 2006 to July 2009 to learn
Cytomegalovirus serum immunoglobulin G antibody
was evaluated using a commercial enzyme-linked im-
Professional Diagnostics) in accordance with the manu-
facturer’s instructions at CCHMC laboratories. During
recruitment, potential participants were told that if eli-
gible (CMV seronegative), they may be invited to par-
ticipate in a CMV vaccine trial, but that their
subsequent participation was voluntary.
To learn more about the epidemiology of CMV,
participants completed a questionnaire that was devel-
oped from previous studies [24–25] and the experience
of the investigators. The questionnaire included ap-
proximately 6 categories of risk: race/ethnicity, age,
exposure to children 3 years of age in the household,
group living situations, saliva-sharing behaviors (in-
cluding kissing), and sexual activity. The questionnaire
was administered using a computer-assisted screening
interview (CASI). Questions regarding sexual activity
were restricted to participants aged 14 years (the legal
age of consent). For this study, race was defined by
participants listing all applicable terms from the fol-
lowing list: African American/black, American Indian/
Alaskan Native, Asian, Hawaiian/Pacific Islander, and
Caucasian/white.Each participant was asked to
describe herself as “Hispanic” or “non Hispanic,” re-
gardless of race. Before study initiation, the institu-
tional review boards of each site approved this study,
and a certificate of confidentiality was obtained. A
parallel seroepidemiology study with the same ques-
tionnaire was performed in adolescent males at
CCHMC, UTMB, and VUMC from June 2006 to
July 2007 as has been previously reported .
Data analyses were performed with SAS version 9.2
(SAS Institute, Cary, NC). Point prevalence calcula-
tions determined the overall seroprevalence of CMV.
To determine factors associated with CMV seropreva-
lence, we first performed univariate χ2analyses for
categorical variables and Wilcoxon tests for continu-
ous variables. Logistic regressions were then per-
formed for multivariate analyses and additive models.
Questions qualifying various exposures as “some-
times,” “often,” or “never or rarely” were collapsed
into “sometimes or often” versus “never or rarely” due
to small sample size. Individuals were then grouped
according to Hispanic, African American (Non-
Hispanic), White (Non-Hispanic), and Other (Non-
Hispanic) categories for the purposes of the analyses.
After the univariate analysis of 6 variables (discussed
above) was performed, risk factors were entered into a
multivariate logistic regression model with a cutoff
value of P< .10.
Of the 1627 adolescent females enrolled in the study,
26 subjects did not complete the questionnaire and
8Stadler et al
another 16 subjects did not have serology results.
Therefore, the number of subjects who completed the
questionnaire (ie, had laboratory results available) and
were included in the statistical analysis was 1585.
There were no significant differences between those in-
cluded in the analysis and those with incomplete data.
Table 1 displays the CMV serostatus of participants
by site. Overall, 779 of the 1585 (49%) participants
were seropositive for CMV. The CMV serostatus
varied significantly across sites (P =0.025); however,
race/ethnicity also varied significantly across sites
(P <.0001). When stratifying the CMV serostatus
comparison by race/ethnicity, the CMV serostatus was
not significantly different across sites (P =.23), and in-
cluding site in the multivariate models did not signifi-
cantly improve the model fit. Because our goal was to
evaluate personal behaviors and racial and ethnic
characteristics, the participant population as a whole
Over half of participants (810) indicated they were
of African American (Non-Hispanic) (51%). Of the
remainder (775), 431 reported being White (Non-
Hispanic) (27%), 214 White (Hispanic) (14%), and
130 reported Other, Non-Hispanic race/ethnic group
(8%). The mean age of the sample was 15.2 years
with a standard deviation of 1.6. Overall, 779 of the
1585 (49%) participants were seropositive for CMV.
CMV antibody was detected in 49% of 12-year-old
participants and ranged from 46% to 51% in those
13–17 years. When a category of age <4 years of age
and 14 years of age was studied, there was no statisti-
cally significant increase in CMV seropositivity with
age (odds ratio [OR], 1.04; confidence interval [CI],
0.83 and 1.30).
Table 2 displays the univariate analysis of the 7 major
risk factors evaluated with regard to CMV serostatus.
Analysis by race/ethnicity revealed that using the refer-
encegroup ofWhite (Non-Hispanic),African
American (Non-Hispanic) descent had over a 2-fold
increased odds ratio (OR, 2.49) of being CMV sero-
positive (CI, 1.95 and 3.17).
Exposure to Young Children in the Home
Of the 1585 participants, 859 (54%) reported chil-
dren< 3 years of age in the home. Of these 859 partic-
ipants, 614 reported 1–2 children in the home,
whereas 245 had> 3 children in the home. Exposure
to a young child was associated with an almost 2-fold
increased odds (OR, 1.88) of CMV infection (CI, 1.54
and 2.30). When specific activities were evaluated
among those with young children in the home, chang-
ing diapers and feeding children were both associated
with increased odds of infection ([OR, 1.54; CI, 1.14
and 2.09] and [OR, 2.02; CI, 1.37 and 2.98],
Group Living Situations
Half of the adolescents studied (793) had previously
participated in a group living situation, defined as
having been an overnight camp counselor for one
week or more (n =65), attended overnight camp (n =
712), stayed overnight in juvenile detention or jail (n
=96), or stayed overnight in a treatment program for
drug, alcohol, or mental problems (n= 50). Overall,
adolescent females who had been in a group living sit-
uation did not have an increased odds of CMV infec-
tion compared with those who had no reported
history of group living (OR, 0.84; CI, 0.69 and 1.02).
Approximately three fourths of adolescents reported
sharing at least 1 item that likely contained saliva;
171 shared toothbrushes, 930 shared lip balm or lip-
stick, and 958 shared drinks. Of females who shared
items, 598 (49%) were CMV positive; there was no
difference in serostatus based on sharing these items
(OR, 1.05; CI, 0.83 and 1.32). Of 1585 adolescents
studied, 1134 (72%) reported kissing another adoles-
cent (male or female). Of those participants, 51%
Table 1. Cytomegalovirus Serostatus of Adolescent Females by Site Locationa
SITEn CMV +AA (Non-Hispanic)White (Non-Hispanic)(Hispanic)Other (Non-Hispanic)Age, Mean, Years
Abbreviations: AA, African American; BCM, Baylor College of Medicine; CCHMC Cincinnati Children’s Hospital Medical Center; CMV,
cytomegalovirus; UTMB, University of Texas Medical Branch in Galveston; VUMC, Vanderbilt University Medical Center.
aData are No. (%) of participants, unless otherwise indicated.
CMV Infections in Adolescent Females9
were CMV seropositive. When kissing alone was eval-
uated, there were no statistically significant differences
with regard to CMV infection between those who re-
ported a history of kissing and those who did not
(OR, 1.22; CI, 0.98 and 1.52). When evaluating the
saliva-sharing category (including sharing items and
kissing), 1412 (89%) of adolescent females engaged in
this behavior and it was not predictive of CMV infec-
tion (OR, 1.25; CI, 0.90 and 1.73).
Of the 1162 adolescent girls > 14 years of age eligible
to answer questions regarding sexual activity, 1150 re-
sponded. Of these, 554 (48%) indicated they had a
history of intimate sexual contact (defined as oral,
anal, and/or vaginal intercourse). Of those who were
sexually experienced, 72 (13%) had a history of anal
intercourse, 336 (61%) had a history of oral inter-
course, and 456 (82%) had a history of vaginal inter-
course. Vaginal intercourse was associated with CMV
infection in the univariate analysis (OR, 1.30; CI,
1.02 and 1.65). Of those with a history of any type of
sexual contact, 292 (53%) were CMV positive (OR,
1.31; CI, 1.04 and 1.65) compared with those partici-
pants without a history of these sexual activities.
Further evaluation of factors within the category of
sexual activity revealed that the age of first sexual
contact was not associated with CMV antibody,
regardless of whether it was evaluated as an ordinal
value (P =.67 using 2-sided Wilcoxon test) or categor-
ically using an age <14 years vs 14 years variable
(OR, 1.14; CI, 0.76 and 1.72).
Neither the time from first intimate sexual contact
to participation in the study (1.54 years +/– 1.3 years,
Table 2. CMV Serology Results by Risk Factor
Risk FactorCMV Negative n (%) CMV Positive n (%)All Subjects n Odds Ratio 95% CI for Odds Ratio
Age <14 years
Children<3 years in home
No children<3 years in home
Did not change diapers
Did not feed children
Changed diapers or fed children
Did not change diapers or feed
Group living situation
No group living situation
Sharing items containing saliva
No sharing items
Kissed male or female
No history of kissing
Saliva sharing behaviors
No saliva sharing behaviors
No sexual activity
1st sexual activity>14 years
Sexual activity <14 years
History of vaginal intercourse
No history of vaginal intercourse
1.04(0.83, 1.30) .74
1.88 (1.54, 2.30).0001
1.54 (1.14, 2.09).0054
2.02 (1.37, 2.98).0004
0.84 (0.69, 1.02).08
1.05(0.83, 1.32) .71
1.22(0.98, 1.52) .07
1.25(0.90, 1.73) .18
1.31 (1.04, 1.65).02
1.14(0.76, 1.72) .53
1.30(1.02, 1.65) .03
Boldfaced values indicate P< .05.
Abbreviations: AA, African American; CI, confidence interval; CMV, cytomegalovirus.
aWith regard to the reference group of white (Non-Hispanic).
10 Stadler et al
P =.84) or the reported number of vaginal sexual
partners (mean 3.45 +/–4.25, P =.82) were predictive
of CMV antibody.
Using a cutoff of P <.10, 5 predictors (race/ethnicity,
feeding and diapering children <3 years of age in the
home, group living, kissing, and sexual activity) were
entered in the multivariate logistic regression. Table 3
displays the results of the final multivariate model, in-
cluding sexual activity (n=1116), in which only race/
ethnicity (P< .0001) and diapering and feeding young
children in the home (P = .0016) remained predictors
of CMV antibody. Being African American (Non-
Hispanic) was associated with 2.49-times increased
odds of CMV seropositivity (CI, 1.83 and 3.38) com-
pared with White (Non-Hispanic), whereas diapering
and feeding children in the home was associated with
1.24-times increased odds of CMV infection (CI, 1.08
When including those 12–13 years of age in the
sample of participants, which meant excluding sexual
activity from the model (n=1531), race/ethnicity was
still associated with a 2-fold (2.01-times) increased
odds of CMV infection (CI, 1.55 and 2.61), and
feeding and diapering children in the home was associ-
ated with 1.3-times increased odds of CMV infection
(CI, 1.16 and 1.45). No other significant predictors
Infection with CMV is very common in adolescent
females, our study showed that 49% of teenage girls
was seropositive. This result is comparable to previ-
ously published US estimates in this age group (41%–
89%) [5, 6, 23, 25–26]. The fact that roughly half of
this population is already infected with CMV should
be taken into account when determining the optimal
target age for a CMV vaccine. Therefore, the age of
optimal vaccination may be much earlier in life.
African Americans and Hispanics have been previ-
ously reported to have increased rates of positive
CMV antibody (in both males and females) compared
with Caucasians [2, 3, 5, 27–29]. Our study confirmed
higher rates in African American (Non-Hispanics)
females compared with other groups, which is similar
to Staras et al’s  National Health and Nutrition
Examination Survey sample from 1988 to 1994 and
Bate et al’s  1988–2004 population. It is of interest
that we were not able to confirm the results of a
smaller study conducted by Wilms et al , who de-
tected a much lower seroprevalence (ranging from
22% to 33% in 13–20 year olds) in African
Americans in 2005–2006 in Virginia. It is unclear
whether factors such as sample size, gender, geograph-
ic location, and sexual activity may be related to the
lower seroprevalence found in that study.
It is likely that exposures to and behaviors associat-
ed with body fluids known to transmit CMV contrib-
ute to the risk of infection. CMV transmission has
traditionally thought to be bimodal with acquisition
occurring either in infancy (secondary to breast
feeding) and/or early childhood (as documented by
day care studies) [24, 31–34] or later, in young adult-
hood (likely due to intimate exposures because CMV
has been detected in saliva [22, 24], cervical specimens
[18, 24], and semen [17, 24]). Many studies in young
women have shown a relationship between sexual ac-
tivity and CMV infection with regard to early sexual
debut [2, 32], heterosexual contact [19, 24], increased
number of sexual partners , and a history of sexu-
ally transmitted infections . We found a relation-
ship (OR, 1.31; CI, 1.04 and 1.65) between sexual
exposure and CMV infection in the univariate, but
not in the multivariate analysis (OR, 1.13; CI, 0.864
and 1.47). Further evaluation of the type(s) of inter-
course, number of sexual partners, and sexually trans-
associations with CMV. It appears that exposure to
young children is a more important source of CMV
infection than sexual activities in adolescents.
When evaluating intimate behaviors in adolescents,
sharing of items that likely contain saliva was com-
monly reported (77%) as was kissing (72%). It is in-
teresting that the use of saliva-sharing items, kissing
and saliva-sharing behaviors, were not linked to CMV
infection because CMV is commonly detected in this
reveal any additional
Table 3. Multivariate Analysis
Risk FactorOdds Ratio95% CI for Odds Ratio
Boldfaced values indicate P<.05.
Abbreviations: AA, African American; CI, confidence interval.
CMV Infections in Adolescent Females 11
body fluid. However, it is problematic to know how
to ask about the frequency and/or intensity of kissing
to explore this possible risk factor. In addition, the ex-
isting literature has documented that living in crowded
situations (in a communal Kibbutz living situation)
and group living has been associated with CMV infec-
tion [5, 35–38]; however, after studying specific group
living situations individually and collectively as a vari-
able, we found that it was not associated with CMV
infection. While some studies have evaluated the
number of individuals living in a home as a measure
of “crowding” or “group living,” we were concerned
that adolescents answering a CASI may not know of
the number of individuals living in the home earlier in
life (from birth to the time of the questionnaire). It is
possible that these variables were not of any signifi-
cance in that we did not sufficiently capture the per-
sonal behaviors or exposure that is linked before
CMV infection. A limitation of a cross-sectional sero-
prevalence study is that we may not be able to detect
the specific behavior or exposure that a child may
have had much earlier in life. A risk factor study at a
younger age (before CMV infection) or a prospective
design study may be more helpful in determining the
specific risks linked with CMV antibody.
In the final multivariate analyses, only African
Americans feeding/diapering young children in the
home were significantly associated with CMV infec-
tion in adolescent females. We previously reported
on a parallel study of adolescent males from the
same clinic populations from Cincinnati, Galveston,
and Nashville . The findings were remarkably
similar; overall, the CMV seroprevalence of adoles-
cent males was 47% and that of adolescent females
49%, which is also similar to the most recent US es-
timates [5, 6]. In addition, the results of both studies
indicated that African Americans were at increased
risk for CMV infection. It is not clear why race was
associated with CMV infection in these clinic popula-
tions, and it may be a marker for some other behav-
ioral or cultural factor. In adolescent males, age was
associated with CMV infection, although it was not
in adolescent females. For both males and females,
exposure to children 3 years in the home was associ-
ated withinfection in
however, it did not remain in the multivariate model
for the adolescent males. Perhaps adolescent women
participate in the care of the children (by feeding and
changing diapers and thus possibly being exposed to
saliva and urine, respectively) more than adolescent
There are a number of limitations to this study, in-
cluding a sample size that may have been too small to
detect subtle differences between groups or in the in
the number or amount of specific behaviors and expo-
sures. Participants were also recruited in conjunction
with a CMV vaccine trial, and although adolescent
females were not required to have an interest in this
study, it may have skewed participation towards those
individuals who were interested in receiving the CMV
vaccine. There may be differences in adolescent
females willing to complete a questionnaire versus
those who are not willing to do so. Therefore, these
participants may have different characteristics than
those attending the adolescent clinics. In addition,
there may have been a cohort effect as well as geo-
graphic differences that may have affected our results,
and thus our findings may not be generalizable of
those living in other geographic locations. It is also
possible that adolescents had errors in self-reporting
behaviors and exposures. In the study, only 1162 of
1585 participants were 14 years of age (and thus
allowed to answer questions pertaining to sexual ac-
tivity), reducing the ability to assess sexual activity as
a risk factor and eliminating the possibility to evaluate
those individuals under the age of consent. In general,
it is difficult to assess possible lifetime exposures using
Specifically, factors relating to the participant’s gesta-
tional period, neonatal period, and early childhood
daycare) were not evaluated in an effort to focus on
adolescent exposures and personal behaviors. In addi-
tion, we were concerned that participants’ answers re-
garding this early information may be unknown,
unavailable, or unreliable. Regardless of the question,
adolescents may error (whether intentionally or unin-
tentionally) in the self-report process. Although the
study adds to the current understanding of CMV in-
fections of adolescent females in the United States, its
cross-sectional design limits the ability to determine
the annual incidence in the population in a prospective
fashion. Given that half of adolescent females have
already been infected with CMV, future studies should
prospectively evaluate exposures and personal behav-
iors in younger children in preparation for a vaccine
aimed at preventing congenital CMV infection.
design in adolescence.
We thank Richard Ward, PhD, and Jesse Lepage for per-
forming the CMV antibody assays (CCHMC); study
12 Stadler et al
coordinators Coni Cheesman and Celsa Tajonera (BCM),
Tara Foltz, Danielle Shoreman (CCHMC), Olivia Doherty,
Gerianne Casey, and Marianne Shafer and Karen Waterman
(UTMB) for their efforts in recruitment and sound advice;
Richard Rupp, MD, and Lawrence R. Stanberry, MD, PhD,
for help in designing the study and the implementation
(UTMB). We also thank Emily Foster, Shanda Phillips,
Belinda Johnson, Jennifer Kissner, PhD, Lisa Sherden,
Matthew Urmy, and Mary Vozar for efforts in study imple-
mentation and subject recruitment at VUMC. Finally, we
thank Heather Hill and Jill Barrett (EMMES) for help in de-
veloping the CASI.
Financial support. This work was supported by the
National Institute of Allergy and Infectious Diseases (grants
N01-AI-25459 and AI-80006 to Cincinnati Children’s
Hospital, grant N01-AI-8007 to Vanderbilt University, and
grants N01-AI-25465 and N01-AI-80002 to Baylor College
of Medicine); Molecular Epidemiology Child Environmental
Health-National Institute of Environmental Health Sciences
(training grant 5-T32-ES010957-08 to L. P. S.); and the
National Institutes of Health (loan repayment extramural
grant funding 2L30AI066732-02 for clinical research from
October 2005 through July 2008 to L. P. S.).
Potential conflicts of interest. K. M. E. receives contract
support from the National Institutes of Health and the Centers
for Disease Control and Prevention. F. M. M. is on the speak-
er’s bureau for Sanofi Pasteur and receives research support
from the National Institutes of Health. C. B. T. participates in
research with Merck for pneumococcal vaccines and Pfizer for
human papillomavirus, meningococcal, and Tdap vaccines,
and receives contract support from the National Institutes of
Health. All other authors report no potential conflicts.
All authorshave submitted
Disclosure of Potential Conflicts of Interest. Conflicts that
the editors consider relevant to the content of the manu-
script have been disclosed.
the ICMJE Formfor
1. Kumar ML, Nankervis GA, Cooper AR, et al. Primary
Pediatrics 1984; 74:493–500.
2. Chandler SH, Holmes KK, Wentworth BB, et al. The
epidemiology of cytomegaloviral infection in women at-
tending a sexually transmitted clinic. J Infect Dis 1985;
3. Sohn YM, Oh MK, Balcarek KB, et al. Cytomegalovirus
infection in sexually active adolescents. J Infect Dis 1991;
4. Nesmith JD, Pass RF. Cytomegalovirus infection in ado-
lescents. Adolesc Med 1995; 6: 79–90.
in adolescent pregnancy.
5. Staras SA, Dollard SC, Radford KW, et al. Seroprevalence
of cytomegalovirus infection in the United States, 1988–
1994. Clin Infect Dis 2006; 43:1143–51.
6. Bate SL, Dollard SC, Cannon MJ. Cytomegalovirus
seroprevalence in the United States: the National Health
and Nutrition Examination Surveys, 1988–2004. Clin
Infect Dis 2010; 50:1439–47.
7. Cannon MJ, KF Davis. Washing our hands of the con-
genital cytomegalovirus disease epidemic. BMC Public
Health 2005; 5:70.
8. Stagno S, Pass RF, Dworsky ME, et al. Maternal cyto-
megalovirus infection and perinatal transmission. Clin
Obstet Gynecol 1982; 25:563–76.
9. Fowler KB, McCollister FP, Dahle AJ, et al. Progressive
and fluctuating sensorineural hearing loss in children
with asymptomatic congenital cytomegalovirus infec-
tion. J Pediatr 1997; 130:624–30.
10. Fowler KB, Stagno S, Pass RF, et al. The outcome
of congenital cytomegalovirus infection in relation to ma-
ternal antibody status. N Engl J Med 1992; 326:663–7.
11. Dobbins JG, Steward JA, Demmler GJ. Surveillance of
Collaborating Registry Group. MMWR CDC Surveill
Summ 1992; 41:35–9.
12. Stratton D, Lawrence RS, eds. Vaccines for the 21st
Century: A Tool for Decision Making. Washington, DC:
National Academy Press; 2000.
Cytomegalovirus glycoprotein-B vaccine with MF59
adjuvant in transplant recipients: a phase 2 randomised
placebo-controlled trial. Lancet 2011; 377:1256–63.
14. Kharfan-Dabaja MA, Boeckh M, Wilck MB, et al. A
novel therapeutic cytomegalovirus DNA vaccine in allo-
geneic haemopoietic stem-cell transplantation: a rando-
mised, double-blind, placebo-controlled, phase 2 trial.
Lancet Infect Dis 2012; 12:290–9.
15. Pass RF, Zhang C, Evans A, et al. Vaccine prevention
of maternal cytomegalovirus infection. N Engl J Med
16. Sung H, Schleiss MR. Update on the current status of cyto-
megalovirus vaccines. Expert Rev Vaccines 2010; 9:1303–14.
17. Lang DJ, JF Kummer. Demonstration of cytomegalovi-
rus in semen. N Engl J Med 1972; 287:756–8.
18. Jordan MC,Rousseau
Association of cervical cytomegalovirus infection with
venereal disease. N Engl J Med 1973; 288:932–4.
19. Handsfield HH, Chandler SH, Caine VA, et al.
Cytomegalovirus infection in sex partners: evidence for
sexual transmission. J Infect Dis 1985; 151:344–8.
20. Pass RF, Hutto C, Ricks R, et al. Increased rate of cyto-
megalovirus infection among parents of children attend-
ing day-care centers. N Engl J Med 1986; 314:1414–8.
21. BaleJF, Zimmerman
Cytomegalovirus transmission in child care homes. Arch
Pediatr Adolesc Med 1999; 153:75–9.
22. Zanghellini F, Boppana
Asymptomatic primary cytomegalovirus infection: virologic
and immunologic features. J Infect Dis 1999; 180:702–7.
23. Stanberry LR, Rosenthal SL, Mills L, et al. Longitudinal
risk of herpes simplex virus (HSV) type 1, HSV type 2,
A,McCarrell E, et al.
CMV Infections in Adolescent Females13
and cytomegalovirus infections among young adolescent Download full-text
girls. Clin Infect Dis 2004; 39:1433–8.
24. Cannon MJ, Hyde TB, Schmid DS. Review of cytomeg-
alovirus shedding in bodily fluids and relevance to con-
genital cytomegalovirus infection. Rev Med Virol 2011;
25. Cannon MJ, Westbrook K, Levis D, et al. Awareness of
and behaviors related to child-to-mother transmission
of cytomegalovirus. Prev Med 2012; 54:351–7.
26. Stadler LP, Bernstein DI, Callahan ST, et al. Seroprevalence
of cytomegalovirus (CMV) and risk factors for infection in
adolescent males. Clin Infect Dis 2010; 51:76–81.
27. Rosenthal SL, Stanberry LR, Biro FM, et al. Seropre-
valence of herpes simplex virus types 1 and 2 and cytomeg-
alovirus in adolescents. Clin Infect Dis 1997; 24:135–9.
28. Tookey PA, Ades AE, Peckham CS. Cytomegalovirus
prevalence in pregnant women: the influence of parity.
Arch Dis Child 1992; 67:779–83.
29. Staras SA, Flanders WD, Dollard SC, et al. Influence of
sexual activity on cytomegalovirus seroprevalence in the
United States, 1988–1994. Sex Transm Dis 2008; 35:
30. Wilms IR, Best AM, Adler SP. Cytomegalovirus Infections
among African Americans. BMC Infect Dis 2008; 8:107.
31. Pass RF, Hutto C. Group day care and cytomegaloviral
infections of mothers and children. Rev Infect Dis 1986;
32. Fowler KB, RF Pass. Risk factors for congenital cyto-
megalovirus infection in the offspring of young women:
exposure to young children and recent onset of sexual
activity. Pediatrics 2006; 118:286–92.
33. Adler SP,FinneyJW,
Prevention of child-to-mother transmission of cytomeg-
alovirus among pregnant women. J Pediatr 2004; 145:
34. Adler SP. Cytomegalovirus transmission among children
in day care, their mothers and caretakers. Pediatr Infect
Dis J 1988; 7:279–85.
35. Sarov B, Naggan L, Rosenzveig R, et al. Prevalence of
antibodies to human cytomegalovirus in urban, kibbutz,
and Bedouin children in southern Israel. J Med Virol
36. Clarke LM, Duerr A, Feldman J, et al. Factors associat-
ed with cytomegalovirus infection among human immu-
nodeficiency virus type 1-seronegative and seropositive
women from an urban minority community. J Infect
Dis 1996; 173:77–82.
37. Palacios O, Cabau N, Horaud F, et al. Serologic survey
of antibodies to cytomegalovirus in women and infants
in Lima, Peru. J Infect Dis 1983; 147:777.
38. Kothari A, Ramachandran VG,
Seroprevalence of cytomegalovirus among voluntary
blood donors in Delhi, India. J Health Popul Nutr
Gupta P, et al.
14 Stadler et al