Wisconsin. Received April 23, 2014; accepted June 18, 2014; electronically published July 17, 2014.

Transcription

1 Original Article Coprevalence of Epstein-Barr Virus, Cytomegalovirus, and Herpes Simplex Virus Type-1 Antibodies Among United States Children and Factors Associated With Their Acquisition Amanda S. Delaney, 1,a William Thomas, 2 and Henry H. Balfour Jr 3,4 Divisions of 1 Epidemiology and Community Health; 2 Biostatistics, School of Public Health, University of Minnesota, Minneapolis; Departments of 3 Laboratory Medicine and Pathology; and 4 Pediatrics, University of Minnesota Medical School, Minneapolis Corresponding Author: Henry H. Balfour Jr, MD, University of Minnesota, Department of Laboratory Medicine and Pathology, MMC 437 Mayo, 420 Delaware Street, Minneapolis, MN balfo001@umn.edu. a Present Affiliation: MD Candidate, Class of 2016, University of Wisconsin School of Medicine and Public Health, Madison, Wisconsin. Received April 23, 2014; accepted June 18, 2014; electronically published July 17, Background. Epstein-Barr virus (EBV), cytomegalovirus (CMV), and herpes simplex virus type-1 (HSV-1) infections are common worldwide, but age-specific prevalence of primary infection varies by race or ethnicity and geographical location. Comparing demographic groups could identify factors influencing the rate of acquisition, age-specific antibody prevalence is relevant for determining when to administer prophylactic vaccines, and coprevalence suggests similar risk factors. Methods. Stored sera collected from the cross-sectional National Health and Nutrition Examination Surveys cycle were tested for EBV, CMV, and HSV-1 antibody. Demographic information was obtained through self-reported questionnaires. Statistical analysis included logistic regression and multivariate analysis adjusting for the multistage cluster design. Results. Overall, 36% of children had antibody against 2 or more of the viruses. Coprevalence with EBV, CMV, and HSV-1 was higher in females, in non-hispanic blacks, and Mexican Americans, compared with non- Hispanic whites, and in those without health insurance. Antibody prevalence was associated with (1) lower household income and education and (2) greater crowding. Nearly all children with CMV antibody or HSV-1 antibody had been infected with EBV. Conclusions. There was a disproportionately high prevalence of EBV, CMV, and HSV-1 antibody among Mexican Americans and non-hispanic blacks, groups with a lower poverty income ratio, and those with less household education. They might benefit from receiving prophylactic herpes vaccines when fairly young. The presence of EBV, CMV, or HSV-1 antibody increases the odds of having antibody against one of the other viruses and is a ripe area for future research. Key words. CMV and HSV-1 antibody coprevalence; cytomegalovirus; EBV; epidemiology of herpesvirus infections; Epstein-Barr virus; herpes simplex virus type 1. Herpesvirus infections are ubiquitous worldwide, but age-specific prevalence of primary infection varies by race or ethnicity and geographical location [1, 2]. These viruses commonly infect children and presumably are spread most often by close contact with infectious individuals. Primary herpesvirus infections may or may not be symptomatic. However, after primary infection, all individuals become lifetime carriers [3]. Herpesvirus infections can lead to serious outcomes. Primary Epstein-Barr virus (EBV) or cytomegalovirus (CMV) infection can manifest as infectious mononucleosis [1, 4]. In immunocompromised individuals, replication of herpesviruses may not be controlled by the immune system. In transplant patients, this can result in posttransplant lymphoproliferative disorder due to EBV or to tissue-invasive CMV disease. Epstein-Barr virus has also been associated Journal of the Pediatric Infectious Diseases Society, Vol. 4, No. 4, pp , DOI: /jpids/piu076 The Author Published by Oxford University Press on behalf of the Pediatric Infectious Diseases Society. All rights reserved. For Permissions, please journals.permissions@oup.com.

2 324 Delaney et al with endemic Burkitt lymphoma, nasopharyngeal carcinoma, Hodgkin disease, multiple sclerosis, and the chronic fatigue syndrome. Cytomegalovirus also causes congenital infections, which can be associated with birth defects especially sensorineural defects. Primary herpes simplex type-1 (HSV-1) infections often manifest as gingivostomatitis [3] in infants and toddlers. Herpes simplex type-1 infections can be life threatening for neonates. Prophylactic vaccines against EBV [5, 6], CMV [7], and HSV-2 [8] are being developed. Because our research program focuses on the pathogenesis of EBV infections, we are most interested in an EBV vaccine and recently completed a study to help determine the best age at which to administer it [2]. That study measured EBV antibody in sera from children 6 19 years old collected by the National Health and Nutrition Examination Surveys (NHANES) between 2003 and We chose NHANES because its multistage probability cluster sample design provides a nationally representative sample of the US noninstitutionalized civilian population. With the EBV antibody data we generated now available for NHANES, the aim of the present study was to examine relationships between EBV, CMV, and HSV-1 antibody prevalence and demographics including age, sex, race or ethnicity, education, crowding, and poverty. This information is necessary to provide policy recommendations about which age groups would benefit most from vaccination to prevent these infections. Our study was limited to the NHANES cycle because that was the only cycle with antibody data for CMV and HSV-1 in addition to EBV. Herpes simplex type-2 and varicella zoster virus (VZV) antibody data were also available, but we chose not to use them because only 18- to 49-year-old subjects were tested for HSV-2 and the presence of VZV antibodies could be confounded by receipt of varicella vaccine. Feeding and diapering children and being black have been found to be associated with CMV antibody prevalence [9]. Crowding [10], daycare attendance [11], socioeconomic status [11, 12], and maternal education [13] have been linked to early acquisition of EBV, but recent studies have not examined these relationships or the coprevalence of the 3 herpesviruses. METHODS Survey Design The National Health and Nutrition Examination Surveys has conducted a series of survey programs since Its goal is to obtain health and nutritional information about the US population, which is used for many different studies. The National Health and Nutrition Examination Surveys samples approximately 5000 noninstitutionalized US civilians every year using a complex multistage probability cluster design strategy to yield a nationally representative sample. The National Health and Nutrition Examination Surveys oversamples specific groups including African-American and Mexican-American individuals. For our study, we used NHANES that sampled 5081 children. The National Health and Nutrition Examination Surveys collected information through household interviews and examinations. A household reference person also completed an interview for each child. The household reference person is the first person listed on the Screener household member roster 18 years or older who owns or rents the child s residence. Participants were eligible for a blood draw unless they had hemophilia, presence of rashes, gauze dressings, casts, edema, paralysis, tubes, open sores or wounds, withered arms or missing limbs, veins that were damaged, sclerosed or occluded, allergies to cleansing agents, burned or scarred tissue, shunts or intravenous lines on both arms, or had received chemotherapy within the last 4 weeks. Surplus sera were available from more than 70% of children (n = 2870) who had a physical examination. We only included those participants aged 6 19 years who had an unequivocal EBV and CMV antibody test result. Herpes simplex virus type-1 antibody was only tested for those aged years. This study was approved by the Research Subjects Protection Program of the University of Minnesota and the National Center for Health Statistics (NCHS) Research Ethics Review Board. Virology Laboratory Procedures Epstein-Barr Virus Antibody Enzyme Immunoassay. Antibodies against EBV viral capsid antigen immunoglobulin (Ig)G were measured using semiquantitative enzyme immunoassay (EIA) kits (Diamedix, Miami, FL). We chose this method because we have tested over 4000 samples in separate 158 runs, with only 3 batch failures ( failure rate, 1.9%). We have published results from 3 studies in which the quantity of the antibody response correlated closely with clinical findings and EBV viral loads [14 16]. In brief, 25 μl of sample are required for each assay. Specimens, calibrators, and controls were prediluted 1:21 before placing them in the test wells. Cutoff calibrator, controls, and subject samples were diluted according to the specific test kit protocol, mixed briefly by vortex to assure good distribution of the sample in the diluent, and then transferred to the appropriately labeled wells and incubated. After washing the wells, conjugate was added with another incubation and wash step after that. Substrate was added and allowed to

3 Herpesvirus Antibodies in US Children 325 incubate before the addition of stop or color reagent. The plates were read at an absorbance of 450 nm using a reference wavelength of 630 nm. Enzyme immunoassay indices were calculated by hand using the formulas in the kit package insert. Data were recorded as Positive (EIA index 1.10), Negative (EIA index <0.90), or Equivocal (EIA index ), and as the exact numeric EIA index. Subjects with equivocal results were not included in the data analysis because we had only 1 sample per subject and therefore could not determine whether the subject was incubating a primary EBV infection and was in the process of seroconverting or had nonspecific reactivity in the EIA antibody assay. Cytomegalovirus Antibody Enzyme Immunoassay. Sera were screened by Bate et al [17] for CMV-specific IgG antibody with the SeraQuestEIA (Quest International) and the Triturus enzyme-linked immunosorbent assay robot (Grifols USA). Samples with reactivities within a narrow range of the SeraQuest EIA cutoff were tested using the VIDAS test (biomérieux Vitek). Their antibody results were used in our analyses. Herpes Simplex Virus Type-Specific Immunodot Antibody Assays. Sera were tested for HSV-1 and -2 antibodies by Xu et al [18] using purified glycoproteins specific for HSV-1 (gg-1) or gg-2 as antigens to detect type-specific antibodies with solid-phase enzymatic immunodot assays. Their antibody results were used in our analyses. Outcome Measures Risk factors examined included age, sex, household reference person s education level (less than high school, high school or GED, or more than high school), crowding ratio (ratio of number of persons per household to number of rooms per household), health insurance (insured or not insured), and race or ethnicity (non-hispanic white, non-hispanic black, or Mexican American). Poverty income ratio was calculated as the household income level divided by poverty threshold set by the US Census Bureau (low [1.3], middle [ ], high [>3.5]). Data Analysis All EBV antibody prevalence estimates were calculated using NCHS exam weights to represent the civilian, noninstitutionalized US population 6 to 19 years old, using SAS Survey Logistic procedure (version 9.3; SAS Institute, Inc. 2010). Prevalence estimates were predictive margins (back-transformed predicted means on the logit scale) from survey logistic regression models fitted to each virus antibody status separately, using each demographic characteristic as independent variable without adjustment. Linear contrasts between fitted prevalence estimates were used to compare factor categories within a factor for each virus. A 2-sided significance level was set at a P value of <.05. SAS Survey Logistic procedure fits logistic regression models for discrete response survey data by using the maximal likelihood method and uses the Taylor series (linearization) method to estimate variance. All P values and confidence intervals are based on these variance estimates. RESULTS The National Health and Nutrition Examination Surveys surveyed 3337 children 6 19 years old. Of these, 2885 children (86%) had antibody measured for at least 1 of the 3 viruses, and 2405 (72%) had antibody measured for at least 2 of the viruses. Antibody data for HSV-1 were not available for children under age 14. The numbers of children with measured antibody status for each virus are given as sample sizes in Table 1. Overall antibody prevalence was 70% for EBV and 38% for CMV in children 6 19 years old; for those years old, HSV-1 antibody prevalence was 35%. Overall, 36% of children had antibody against 2 or more or the viruses. Age-specific antibody prevalence is shown in Figure 1. Coprevalence with EBV, CMV, and HSV-1 was higher in females, in non-hispanic blacks, and Mexican Americans, compared with white non-hispanics, and in those without health insurance; prevalence was associated with lower household income and education and greater crowding (Table 1). Coprevalence rates for each combination of viruses are showninfigure2a with ages 6 13 years, where only CMV and EBV status were available, and shown separately in Figure 2B from ages years. There was an increase in both EBV and CMV antibody prevalence between ages 6 and 8, with stable prevalence from age 8 to 13 (Figure 2A; Table 1). In children 14 19, EBV antibody prevalence increased more with age than either CMV or HSV-1 antibody prevalence, including those with only EBV antibody (Figure 2B). Nearly all children with CMV antibody or HSV-1 antibody had been infected with EBV; the percentage of children with HSV-1 antibody alone decreased to 1.4% by age 19, and the percentage with CMV antibody alone decreased to 4% by age 19 (Figure 2B). Overall, the risk factors for acquiring EBV, CMV, and HSV-1 were the same, whereas the magnitude of each demographic factor varied by virus (Table 2). When crowding was univariately modeled as a continuous variable, greater crowding was associated with greater antibody prevalence for each virus: EBV (P =.001), CMV (P =.003), and HSV-1 (P =.014). The age-adjusted odds of EBV antibody positivity was 2.01 (P =.042) times higher for those with a

4 Table 1. Unadjusted EBV, CMV, and HSV-1 Antibody Prevalence and Coprevalence Rates in Children 6 19 Years Old in in the United States by Demographic Characteristics, Estimated From NHANES Using Survey Weights* 326 Delaney et al Demographic Factor Coprevalence Antibody Coprevalence % (95% CI) EBV EBV Antibody Prevalence % (95% CI) CMV CMV Antibody Prevalence % (95% CI) HSV-1 HSV-1 Antibody Prevalence % (95% CI) Total (31 41) (65 75) (33 42) (31 39) Gender Male (28 38) a (62 72) a (30 40) a (28 37) a Female (34 46) b (66 78) a (35 46) b (33 44) b Age 6 7 years old (14 28) a (46 64) a (22 36) a years old (21 34) a,b (59 73) b (31 42) a,b years old (25 38) b (57 72) b (31 44) b years old (21 33) a,b (56 72) a,b (31 45) b years old (37 54) c (60 75) b,c (30 48) b (27 43) a years old (41 59) c (74 84) c,d (35 50) b (29 45) a years old (42 58) c (78 89) d (34 47) b (31 40) a Race-Ethnicity White, non-hispanic (22 31) a (56 68) a (23 33) a (24 32) a Black, non-hispanic (51 62) b (82 90) b (50 64) b (48 61) b Mexican American (53 64) b (82 89) b (36 47) b (46 58) b Household Income to Poverty Level Ratio Low (1.3) (44 56) a (79 87) a (43 56) a (41 55) a Middle ( ) (25 40) b (62 73) b (27 41) b (26 40) b High (>3.5) (20 28) c (48 64) c (24 32) b (18 29) c Health Insurance Insured (29 39) a (63 73) a (31 40) a (30 38) a Not Insured (45 56) b (75 87) b (44 58) b (40 52) b Highest Household Education Less than High School (55 69) a (81 92) a (53 64)a (46 62) a High School or GED (35 47) b (66 80) b (30 45) b (37 50) b More than High School (22 32) c (57 68) c (26 37) b (22 31) c Crowding Index (persons/ room) Low (<0.5) (16 28) a (53 69) a (16 29) a (18 33) a Middle (0.5 to 1.0) (32 42) b (64 75) b (34 43) b (31 41) b High (>1.0) (47 62) c (75 89) c (48 63) c (33 58) b Abbreviations: CI, confidence interval; CMV, cytomegalovirus; EBV, Epstein-Barr virus; HSV-1, herpes simplex virus type-1; NHANES, National Health and Nutrition Examination Surveys. *Subjects were included in the sample for coprevalence if they had antibody measured for at least 2 viruses. HSV-1 antibody status was not available in children under 14. Each characteristic was modeled individually for each viral outcome. Sample sizes are unweighted counts of children with measurement of viral antibody. Within each viral outcome group, categories within a demographic factor were compared by pairwise linear contrasts. For ages 6 13 years, coprevalence is both EBV and CMV antibody prevalence; for ages years, coprevalence includes children with 2 or more of EBV, CMV, and HSV-1 antibody prevalence. a c Within each racial or ethnic group, categories within a demographic factor were compared by pairwise linear contrasts. Rates with no letters in common were significantly different (P <.05), whereas rates that shared letters were not significantly different.

5 Herpesvirus Antibodies in US Children 327 Figure 1. Age-specific prevalence of antibodies against Epstein-Barr virus (EBV), cytomegalovirus (CMV), and herpes simplex virus type-1 (HSV-1) among 6- to 19-year-old participants in National Health and Nutrition Examination Surveys, cycle. Herpes simplex virus type-1 antibody status was not available in children under 14 years of age. household reference person who did not complete 9th grade compared with those who had a household reference person who completed college or above. Likewise, lower household education level (below 9th grade) was associated with increased odds of being positive for CMV antibody (OR = 1.62, P =.018). This association was even stronger for HSV-1, with an age-adjusted odds ratio of 2.27 (P <.001). Concerning coprevalence of antibodies, those who were EBV antibody positive were more likely to be CMV antibody positive (OR = 2.23, P <.0001) or HSV-1 antibody positive (OR = 4.80, P <.0001) compared with those who were EBV negative (Table 3). The odds of having both CMV and HSV-1 antibodies were times higher (P <.0001) among those who were EBV positive compared with those who were EBV negative (Table 3). When conducting age-adjusted multivariate analysis, significant risk factors for each of the 3 viruses included lower poverty income ratio, less household education, and being black (non-hispanic) or Mexican American (Table 2). Crowding ratio became nonsignificant for each virus after adjusting for other risk factors and was removed from the final model. Although the magnitude of association between risk factors and antibody status was attenuated after adjustment, lower poverty income ratio remained significant. The age-adjusted odds of having EBV antibody was 3.04 (P <.0001) times higher for those with a poverty income ratio 1.3 compared with those whose poverty income ratio was greater than 3.5. The trend for CMV antibody prevalence was in the same direction with an odds ratio of 1.70 (P <.001) and HSV-1 with an odds ratio of 1.83 (P <.0001). DISCUSSION Figure 2. Unadjusted Epstein-Barr virus (EBV), cytomegalovirus (CMV), and herpes simplex virus type-1 (HSV-1) antibody prevalence and coprevalence rates among 6- to 13-year-old (A) and 14- to 19-year-old (B) participants in National Health and Nutrition Examination Surveys, cycle. Herpes simplex virus type-1 antibody status was not available in children under 14 years of age. An important finding of this study was that presence of antibodies for EBV, CMV, or HSV-1 increased the odds of having antibody against one of the other viruses. Children with EBV antibodies were 4.80 times more likely to have HSV-1 antibodies than children without EBV antibodies. The odds of having CMV antibody was 2.23 times higher when comparing EBV antibody-positive with EBV antibody-negative groups. In addition, those who were EBV antibody positive were times more likely to have both CMV and HSV-1 antibodies. We also examined the demographic risk factors associated with acquisition of EBV, CMV, and HSV-1. In mutually adjusted models, risk factors for acquisition of EBV, CMV, and HSV-1 were the following: lower poverty income ratio, less household education, and being black (non-hispanic) or Mexican American. Crowding was not a significant risk factor for any of the 3 viruses in multivariate age-adjusted analysis. Racial or ethnic differences in CMV antibody

6 328 Delaney et al Table 2. Odds Ratios for Mutually Adjusted Model Examining Risk Factors Demographic Factor EBV Odds Ratio P Value CMV Odds Ratio P Value HSV-1 Odds Ratio P Value Sex Female Male 1.00 (reference) 1.00 (reference) 1.00 (reference) Poverty Income Ratio a < < < > (reference) 1.00 (reference) 1.00 (reference) Household Education Level Less than High School <.001 High School or GED More than High School 1.00 (reference) 1.00 (reference) 1.00 (reference) Race/Ethnicity Mexican American 2.57 < < <.001 Non-Hispanic Black 2.68 < < <.0001 Non-Hispanic White 1.00 (reference) 1.00 (reference) 1.00 (reference) Abbreviations: CI, confidence interval; CMV, cytomegalovirus; EBV, Epstein-Barr virus; HSV-1, herpes simplex virus type-1. a Poverty income ratio was the ratio of family income to the family s poverty threshold. Table 3. Odds Ratios for Coinfections According to EBV Antibody Status CMV Odds Ratio a (95% CI) P Value HSV-1 Odds Ratio b (95% CI) P Value CMV and HSV-1 Odds Ratio b (95% CI) P Value EBV Positive 2.23 (1.71, 2.91) 4.80 (3.14, 7.31) (3.09, 34.19) EBV Negative 1.00 (reference) < (reference) < (reference).0001 Abbreviations: CI, confidence interval; CMV, cytomegalovirus; EBV, Epstein-Barr virus; HSV-1, herpes simplex virus type-1. a CMV data included ages 6-19 years old. b HSV-1 data only included individuals years old. prevalence among NHANES subjects have been previously reported by Staras et al [19] andbateetal[17]. When NHANES subjects from 1998 to 2004 were included, less household income, high household crowding, and less household income in addition to race or ethnicity were associated with CMV antibody prevalence [17]. The consequences of being infected with herpesviruses are not trivial. Tarter et al [20] examined data on NHANES subjects enrolled between 1988 and They found that 12- to 16-year-old children with HSV-1 antibodies had impaired reading vocabulary and visuospatial processing compared with children who lacked HSV-1 antibodies. Being positive for HSV-1 antibodies also affected cognition among adults, whereas CMV antibody positivity was associated with cognitive impairment in adults but not in children. The strong association between the presence of EBV and HSV-1 antibodies suggests that the transmission of these viruses may be similar. An alternative hypothesis is that an EBV or HSV-1 infection could make an individual more susceptible to the other virus. A previous study in England and Wales by Morris et al [21] also found a significant association between EBV and HSV-1 in the age groups and (P <.001). There were several risk factors in common for acquiring EBV, CMV, and HSV-1 antibodies. This is plausible because these 3 viruses are presumably spread through close contact with infected individuals [7]. Poverty income ratio was more influential on EBV antibody prevalence than it was for CMV and HSV-1 antibody prevalence, which is consistent with previous studies [9]. Crowding was independently associated with EBV, CMV, and HSV-1, but it was not significant in multivariate analysis. This finding shows that crowding may be surrogate for other factors including poverty income ratio, education level, and race or ethnicity. A strength of our study is that data were derived from a nationally representative sample of US children tested in 2003 and However, our study has some limitations. Due to its cross-sectional design, there was no information on whether participants were exposed to risk factors before or after acquisition of EBV, CMV, or HSV-1 infection. Household income and education levels were self-reported, which may have been influenced by social desirability. Therefore, the association between antibody status and income or education level may have been weakened. Another limitation is that only year olds were tested for HSV-1 antibody, which precluded our ability to assess risk factors for HSV-1 acquisition at younger ages and our ability to compare risk factors for coprevalence of EBV and CMV antibodies in children 6 14 years old. Because this was a secondary analysis, we were not able to ask additional questions about risk factors that have been found to be significant in previous studies such as daycare attendance [10] and geographic location [1]. In a previous study,

7 Herpesvirus Antibodies in US Children 329 Xu et al [18] found that in NHANES , age-specific acquisition of HSV-1 was slow, so there probably is not an upward inflection point for HSV-1. Because the prevalence of EBV antibody was already high in the youngest cohort (67%), it would be beneficial to include children younger than 6 years old in future studies to see how risk factors influence prevalence of EBV, CMV, and HSV-1 antibody at those ages. A prospective study would be able to address temporal associations. Contemporary studies are needed to determine whether antibody prevalence among US children has changed since In conclusion, we found a strong association between EBV and HSV-1 antibody prevalence among children years old. We also determined that there was a disproportionately high prevalence of EBV, CMV, and HSV-1 antibody among groups with lower poverty income ratio, less household education, and among non-hispanic blacks and Mexican Americans. Because of this outcome, a uniform policy would be to vaccinate all children with herpesvirus vaccines when fairly young. The presence of EBV, CMV, or HSV-1 antibody increases the odds of having antibody against one of the other viruses and is a ripe area for future research. Acknowledgments We thank Peter Hannan and Dr. Simon Rosser for valuable suggestions. We also thank David O. Schmeling and Jennifer A. Knight for technical assistance. Financial support. This work was supported by grants from the University of Minnesota International Center for Antiviral Research and Epidemiology, and the University of Minnesota Foundation. Potential conflicts of interest. All authors: No reported conflicts. All authors have submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest. References 1. Hjalgrim H, Friborg J, Melbye M. The epidemiology of EBV and its association with malignant disease. In: Arvin A, Campadelli-Fiume G, Mocarski E, et al. eds. Human Herpesviruses: Biology, Therapy, and Immunoprophylaxis. Cambridge, UK: Cambridge University Press; 2007: pp Balfour HH Jr, Sifakis F, Sliman JA, et al. Age-specific prevalence of Epstein-Barr virus (EBV) infection among children in the United States and factors affecting its acquisition. J Infect Dis 2013; 208: Roizman B, Whitley RJ, Lopez C, eds. The Human Herpesviruses. New York, NY: Raven Press; Henke CE, Kurland LT, Elveback LR. Infectious mononucleosis in Rochester, Minnesota, 1950 through Am J Epidemiol 1973; 98: Balfour HH Jr. Epstein-Barr virus vaccine for the prevention of infectious mononucleosis--and what else? J Infect Dis 2007; 196: Sokal EM, Hoppenbrouwers K, Vandermeulen C, et al. Recombinant gp350 vaccine for infectious mononucleosis: a phase 2, randomized, double-blind, placebo-controlled trial to evaluate the safety, immunogenicity, and efficacy of an Epstein-Barr virus vaccine in healthy young adults. J Infect Dis 2007; 196: Pass RF, Zhang C, Evans A, et al. Vaccine prevention of maternal cytomegalovirus infection. N Engl J Med 2009; 360: Belshe RB, Leone PA, Bernstein DI, et al. Efficacy results of a trial of a herpes simplex vaccine. N Engl J Med 2012; 366: Stadler LP, Bernstein DI, Callahan ST, et al. Seroprevalence and risk factors for cytomegalovirus infections in adolescent females. J Pediatric Infect Dis Soc 2013; 2: Sumaya CV, Henle W, Henle G, et al. Seroepidemiologic study of Epstein-Barr virus infections in a rural community. J Infect Dis 1975; 131: Hesse J, Ibsen KK, Krabbe S, et al. Prevalence of antibodies to Epstein-Barr virus (EBV) in childhood and adolescence in Denmark. Scand J Infect Dis 1983; 15: Henle G, Henle W, Clifford P, et al. Antibodies to Epstein-Barr virus in Burkitt s lymphoma and control groups. J Natl Cancer Inst 1969; 43: Figueira-Silva CM, Pereira FE. Prevalence of Epstein-Barr virus antibodies in healthy children and adolescents in Vitoria, State of Espirito Santo, Brazil. Rev Soc Bras Med Trop 2004; 37: Balfour HH Jr, Holman CJ, Hokanson KM, et al. A prospective clinical study of Epstein-Barr virus and host interactions during acute infectious mononucleosis. J Infect Dis 2005; 192: Balfour HH Jr, Hokanson KM, Schacherer RM, et al. A virologic pilot study of valacyclovir for infectious mononucleosis. J Clin Virol 2007; 39: Balfour HH Jr, Odumade OA, Schmeling DO, et al. Behavioral, virologic, and immunologic factors associated with acquisition and severity of primary Epstein-Barr virus infection in university students. J Infect Dis 2013; 207: Bate SL, Dollard SC, Cannon MJ. Cytomegalovirus seroprevalence in the United States: the national health and nutrition examination surveys, Clin Infect Dis 2010; 50: Xu F, Lee FK, Morrow RA, et al. Seroprevalence of herpes simplex virus type 1 in children in the United States. J Pediatr 2007; 151: Staras SA, Dollard SC, Radford KW, et al. Seroprevalence of cytomegalovirus infection in the United States, Clin Infect Dis 2006; 43: Tarter KD, Simanek AM, Dowd JB, Aiello AE. Persistent viral pathogens and cognitive impairment across the life course in the third national health and nutrition examination survey. J Infect Dis 2014; 209: Morris MC, Edmunds WJ, Hesketh LM, et al. Sero-epidemiological patterns of Epstein-Barr and herpes simplex (HSV-1 and HSV-2) viruses in England and Wales. J Med Virol 2002; 67: