Skip Navigation


American Journal of Epidemiology Advance Access originally published online on August 8, 2006
American Journal of Epidemiology 2006 164(8):733-741; doi:10.1093/aje/kwj270
This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow All Versions of this Article:
164/8/733    most recent
kwj270v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (7)
Right arrowRequest Permissions
Right arrow Disclaimer
Google Scholar
Right arrow Articles by Brown, E. L.
Right arrow Articles by Celum, C.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Brown, E. L.
Right arrow Articles by Celum, C.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?

American Journal of Epidemiology Copyright © 2006 by the Johns Hopkins Bloomberg School of Public Health All rights reserved; printed in U.S.A.

Original Contribution

High Risk of Human Immunodeficiency Virus in Men Who Have Sex with Men with Herpes Simplex Virus Type 2 in the EXPLORE Study

E. L. Brown1, A. Wald1,2,3, J. P. Hughes4, R. A. Morrow3,5, E. Krantz3, K. Mayer6,7, S. Buchbinder8, B. Koblin9 and C. Celum1,2

1 Department of Epidemiology, University of Washington, Seattle, WA
2 Department of Medicine, University of Washington, Seattle, WA
3 Department of Laboratory Medicine, University of Washington, Seattle, WA
4 Department of Biostatistics, University of Washington, Seattle, WA
5 Children's Hospital and Regional Medical Center, Seattle, WA
6 Department of Medicine, Brown University, Providence, RI
7 Fenway Community Health, Boston, MA
8 San Francisco Department of Public Health, San Francisco, CA
9 New York Blood Center, New York, NY

Correspondence to Dr. Connie Celum, Cabrini Medical Tower, 901 Boren Avenue, no. 1600, Seattle, WA 98104 (e-mail: ccelum{at}u.washington.edu).

Received for publication January 19, 2006. Accepted for publication March 31, 2006.


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 References
 
The relation between herpes simplex virus type 2 (HSV-2) and human immunodeficiency virus (HIV) acquisition was evaluated among 4,295 high-risk, HIV-negative men who have sex with men in an intensive behavioral intervention (colloquially referred to as "EXPLORE") study in the United States from 1999 to 2003. Sexual behavior data were obtained by computer-assisted self-interview, and sera were collected semiannually for HIV and HSV-2 serology. HSV-2 infection was classified as "recent incident" (at the first HSV-2 seropositive visit), "remote incident" (within 24 months of the first positive visit), and "prevalent" (for visits >24 months after the first HSV-2 positive visit). Baseline HSV-2 prevalence was 20.3%. HSV-2 incidence was 1.9 (95% confidence interval (CI): 1.6, 2.2) per 100 person-years; significant risk factors were African-American race, unprotected receptive anal intercourse, an HIV-positive male sex partner, and six or more male partners in the prior 6 months. The behavioral intervention did not reduce HSV-2 acquisition (adjusted hazard ratio (HR) = 1.2, 95% CI: 0.9, 1.6). Overall HIV incidence was 1.9 (95% CI: 1.7, 2.2) per 100 person-years. HIV risk was elevated among men who have sex with men with recent incident HSV-2 (adjusted HR = 3.6, 95% CI: 1.7, 7.8), remote incident HSV-2 (adjusted HR = 1.7, 95% CI: 0.8, 3.3), and prevalent HSV-2 (adjusted HR = 1.5, 95% CI: 1.1, 2.1) infection compared with HSV-2 seronegative participants. HIV intervention strategies targeting HSV-2 prevention and suppression among men who have sex with men should be evaluated.

behavior therapy; herpes genitalis; herpesvirus 2, human; HIV; homosexuality, male

Abbreviations: CI, confidence interval; HIV, human immunodeficiency virus; HR, hazard ratio; HSV-1, herpes simplex virus type 1; HSV-2, herpes simplex virus type 2


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 References
 
Genital herpes, most commonly caused by herpes simplex virus type 2 (HSV-2), is among the most prevalent sexually transmitted infections globally (1Go–3Go). Genital HSV-2 infection increases the risk of human immunodeficiency virus (HIV) type 1 acquisition (4Go–8Go), likely via mucosal or epithelial disruption, which provides a portal of entry for HIV, and via recruitment of CD4-positive T lymphocytes during HSV-2 reactivation (9Go). Because the period immediately following genital HSV-2 acquisition is marked by increased frequency and duration of reactivations (10Go), people with early HSV-2 infection may have an even higher risk of HIV acquisition than do those with established HSV-2 infection.

The EXPLORE study was a randomized, controlled, phase IIb efficacy trial of a 10-session behavioral intervention designed to reduce HIV acquisition among 4,295 high-risk, HIV-negative men who have sex with men. Although there was no significant reduction in HIV incidence (intervention effect = 15.7 percent, 95 percent confidence interval (CI): –8.4, 34.4), a statistically significant reduction of 20.5 percent (95 percent CI: 10.9, 29.0) in unprotected receptive anal intercourse with an HIV-positive or unknown serostatus partner was detected in the behavioral intervention arm (11Go).

We used blood samples and behavioral data collected during the EXPLORE study to assess risk factors for HSV-2 acquisition, to evaluate the association of prevalent and incident HSV-2 infection with HIV acquisition, and to determine the effect of the behavioral intervention on HSV-2 acquisition.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 References
 
The EXPLORE study was conducted from 1999 to 2003 in six US cities: New York, New York; San Francisco, California; Seattle, Washington; Boston, Massachusetts; Chicago, Illinois; and Denver, Colorado. Details of the eligibility criteria, study protocol, primary results, and baseline characteristics of the participants have been reported previously (www.explorestudy.org) (11Go–13Go). Institutional review boards at each study site approved the study. All participants provided written informed consent that included HIV and HSV-2 serologic testing.

Subjects
Participants were recruited between January 1999 and February 2001, as described previously (11Go, 13Go). Men were eligible if they were HIV seronegative, were aged 16 years or older, reported engaging in anal intercourse with at least one male partner in the past year, and were not in a mutually monogamous relationship with an HIV-negative partner lasting 2 or more years.

Study procedures
At enrollment, participants were randomized to either the intervention arm or the control arm. Those in the intervention arm received 10 individualized counseling sessions of 1-hour duration with specific themes delivered over 4–6 months, followed by quarterly maintenance sessions (12Go). Those in the control arm received semiannual HIV testing with pre- and posttest counseling based on the Centers for Disease Control and Prevention's Project RESPECT model (14Go). Sexual risk behavior data were collected semiannually via audio computer-assisted self-interview to increase the accuracy of reporting of sensitive behaviors, such as unprotected sex (15Go–17Go). Blood samples were collected semiannually for HIV testing, with samples stored for future batch testing for HSV-2 antibodies.

Laboratory methods
HSV-2 serologic testing was performed with the commercially available, type-specific HerpeSelect type 2 enzyme-linked immunosorbent assay (Focus Diagnostics, Cypress, California), with herpes simplex virus Western blot (18Go) confirmation of sera with index values between 0.9 and 3.5 because of the lower specificity of the enzyme-linked immunosorbent assay in the low positive (index values = 1.1–3.5) range (19Go, 20Go). HSV-2 testing was performed at the University of Washington after the study was completed. Final visit sera from consenting participants who had stored samples available were tested first, followed by testing of baseline sera of those identified as HSV-2 seropositive at the final visit. Those participants who appeared to seroconvert to HSV-2 during follow-up (based on a negative baseline result and a positive final visit result) had HSV-2 testing performed on all available interim sera, to determine the time of seroconversion. Results were provided to study sites and offered to participants.

Statistical methods
All behavioral and sexually transmitted infection data were reported at each visit for the 6-month period prior to that visit, including diagnosis or treatment for syphilis, gonorrhea, or chlamydia since the prior study visit. At each visit, participants were asked the following questions: "Since your last interview, have you had a recurrent episode or been newly diagnosed with penile or rectal herpes?" and "Have you had sores on your penis or in or around your rectum since your last interview?" Participants answering "yes" to either question were categorized as having genital ulcer disease at that visit. The number of male sex partners since the prior study visit was dichotomized at the median (≤5 vs. ≥6 partners). Serodiscordant intercourse was defined as intercourse with an HIV-positive partner or a partner of unknown HIV status. The number of unprotected receptive anal sex acts was categorized in quartiles. Alcohol use and drug use were categorized as described previously (21Go), and depression was evaluated using a shortened seven-item version of the Center for Epidemiologic Studies Depression Scale (22Go), with a score of at least 13 of 28 considered as reflecting some depressive symptoms. Missing behavioral data were left as missing rather than being carried forward or imputed.

Participants who acquired HSV-2 during follow-up were classified as HSV-2 negative until the first visit at which they tested positive for HSV-2 antibodies. HSV-2 seroconverters were classified as having "recent incident" HSV-2 at the first HSV-2 seropositive visit, "remote incident" HSV-2 for all visits occurring within 24 months after the first positive visit, and "prevalent HSV-2" for all visits occurring more than 24 months after the first positive visit. Thus, a participant who acquired HSV-2 during follow-up could contribute person-time to multiple categories of HSV-2 status. Participants seropositive for HSV-2 at baseline were classified as having prevalent infection at all study visits. Ten participants who acquired HSV-2 during follow-up but were missing interim sera reported genital ulcers during follow-up. For these 10 participants, the timing of HSV-2 acquisition was assigned as the first visit with missing serology when genital ulcers were reported. An additional 27 participants missing interim sera who seroconverted to HSV-2 did not report any genital ulcers. In analyses in which HSV-2 acquisition was the outcome, the time of HSV-2 acquisition for these 27 participants was estimated as the midpoint between the last negative and the first positive HSV-2 test. In analyses in which HSV-2 was a predictor (e.g., time to HIV infection), we imputed the timing of HSV-2 acquisition, assuming an equal probability of HSV-2 acquisition at each study visit for these 27 participants. Given the less precise timing of HSV-2 acquisition in these 37 participants, we conducted sensitivity analyses, assuming extreme values of HSV-2 acquisition time for these 37 participants.

HSV-2 acquisition rates were compared between the intervention and control groups by intention-to-treat analysis. A binomial regression model with log link (23Go) was used to compute per-contact rates of HSV-2 acquisition for receptive, insertive, protected, and unprotected anal intercourse and by partner's HIV status. The model assumes that the per-contact infectivity is independent of the number of sex acts of an individual, and so the per-contact rates should be the same for people who have a high number of sex acts versus low numbers of sex acts. To assess the validity of the model assumption, we repeated these calculations, successively excluding greater proportions of observations containing the highest numbers of reported anal sex acts to determine whether the calculated per-act infectivity remained constant regardless of the number of reported sex acts.

Cox proportional hazards regression models were used to assess the effect of the intervention on HSV-2 acquisition rates, the factors associated with HSV-2 acquisition, and the associations of recent and remote incident and prevalent HSV-2 infection with HIV acquisition. Predictors of HSV-2 acquisition were identified by use of univariate regression; those factors found to be significant in the univariate models (p < 0.1) were included in a multivariate model. Those factors found to remain significant (p < 0.05) in the multivariate model were retained in the final model, which also included study site, study arm, and age.

For the analysis of the associations of incident and prevalent HSV-2 with HIV acquisition, all behavioral covariates found to be predictors of HIV infection in earlier analyses of this data set (21Go) or thought to be important markers of sexual risk behavior were retained in the multivariate model, even though some of these did not alter the risk estimate appreciably. Stata, version 8.0, software (StataCorp LP, College Station, Texas) was used for all analyses.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 References
 
A total of 4,295 HIV-negative men who have sex with men participated in the EXPLORE study. Of these men, 3,909 (91 percent) had valid HSV-2 test results. The remainder of the participants either did not have stored sera available for HSV-2 testing (n = 358) or had indeterminate HSV-2 test results even after the Western blot was performed (n = 28), and they were excluded from the analyses. Of these 3,909 men, 1,980 were in the control arm, with a median follow-up time of 36.2 months (range: 1.2–51.2 months), and 1,929 were in the intervention arm, with a median follow-up of 36.1 months (range: 2.1–52.7 months). The median age was 33 years (interquartile range: 27–39 years); 27 percent were non-Caucasian (14 percent Hispanic, 7 percent African American, and 6 percent other); and the median educational level was a college degree.

Incidence of HSV-2
Among the 3,909 participants with HSV-2 results, 793 (20.3 percent) were HSV-2 seropositive at enrollment, 2,949 (75.4 percent) remained seronegative throughout the study, and 167 (4.3 percent) acquired HSV-2 during follow-up. There was no difference in HSV-2 incidence between the study arms (figure 1). Of the participants who acquired HSV-2, 87 were in the 10-session behavioral intervention arm, and 80 were in the brief risk-reduction counseling control arm (adjusted hazard ratio (HR) = 1.2, 95 percent CI: 0.9, 1.6) (table 1).


Figure 1
View larger version (10K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		FIGURE 1. Kaplan-Meier curve of proportion of EXPLORE study participants in six US cities remaining herpes simplex virus type 2 (HSV-2) uninfected, by study arm, 1999–2003.

 

View this table:
[in this window]
[in a new window]

  		TABLE 1. Factors associated with herpes simplex virus type 2 acquisition among men who have sex with men in six US cities participating in the EXPLORE study, 1999–2003

 
The overall HSV-2 incidence rate was 1.9 (95 percent CI: 1.6, 2.2) per 100 person-years. HSV-2 acquisition rates per 10,000 sex acts with partners of unknown HSV-2 status ranged from 2.4 (95 percent CI: 1.8, 3.1) for all anal intercourse with HIV-negative partners to 5.0 (95 percent CI: 2.7, 7.4) for all anal intercourse with HIV-positive partners. Rates were somewhat higher for all receptive acts (3.9, 95 percent CI: 2.8, 4.9) than for all insertive acts (2.7, 95 percent CI: 1.9, 3.6) and for protected acts (4.2, 95 percent CI: 3.2, 5.2) than for unprotected acts (2.4, 95 percent CI: 1.6, 3.1). In an analysis of the validity of the model assumptions of independence of infectivity of individual sex acts, calculated per contact, HSV-2 acquisition rates varied with the number of sex acts reported, with exclusion of observations with the highest reported numbers of sex acts causing the per-contact rates to increase; for example, if the observations with the highest 50 percent of observations are excluded, then the per-contact rates increase approximately threefold. No data on the HSV-2 serostatus of partners or the distribution of sex acts across partners were available.

Among HSV-2 seroconverters who had their timing of HSV-2 acquisition documented serologically, only 8.1 percent reported a first episode of genital ulcers in the 6-month interval just prior to the first positive HSV-2 test. Genital ulcers within the prior 6 months were reported at some point during the study by 31 percent of the participants who were HSV-2 seropositive at baseline, 32 percent of HSV-2 seroconverters, and 17 percent of participants who remained HSV-2 seronegative. A diagnosis of syphilis in the prior 6 months was reported at 0.34 percent of all study visits and at 1.4 percent of visits at which participants reported genital ulcer symptoms but were HSV-2 seronegative.

Risk factors for HSV-2 acquisition
Factors associated with HSV-2 acquisition in the multivariate model included unprotected receptive anal intercourse (for ≥5 unprotected receptive anal intercourse acts vs. no receptive anal intercourse within the prior 6 months: adjusted HR = 2.6, 95 percent CI: 1.6, 4.1), having at least one HIV-positive male partner within the past 6 months (adjusted HR = 1.6, 95 percent CI: 1.2, 2.3), having six or more male partners in the prior 6 months (adjusted HR = 1.5, 95 percent CI: 1.1, 2.1), and African-American compared with Caucasian race (adjusted HR = 1.9, 95 percent CI: 1.1, 3.2) (table 1). These findings were not sensitive to the timing assumptions made for the 37 participants with uncertain dates of HSV-2 acquisition, because the hazard ratios for these associations do not differ meaningfully as the timing of HSV-2 acquisition is varied from the beginning to the end of the follow-up period for these 37 participants (data not shown).

Association of HSV-2 with HIV acquisition
The overall HIV acquisition rate among this subset (91 percent) of the EXPLORE cohort was 1.9 (95 percent CI: 1.7, 2.2) per 100 person-years. Among the 20 participants who acquired both HSV-2 and HIV during follow-up, HSV-2 was detected before HIV for seven participants. The interval by which HSV-2 preceded HIV ranged from 21 days to 751 days (median: 478 days). For the other 13 participants, HSV-2 and HIV were first detected at the same study visit, and for eight of these 13, no interim sera were available for testing. HSV-2 acquisition was never detected after HIV acquisition, because participants exited the study when they acquired HIV.

HIV acquisition rates were elevated among participants with recent incident HSV-2 (6.9 per 100 person-years, 95 percent CI: 3.1, 15.4), remote incident HSV-2 (6.8 per 100 person-years, 95 percent CI: 3.9, 12.0), and prevalent HSV-2 (2.7 per 100 person-years, 95 percent CI: 2.2, 3.5) compared with the rate among participants who remained HSV-2 seronegative (1.5 per 100 person-years, 95 percent CI: 1.3, 1.8) (figure 2). The unadjusted hazard ratio of HIV acquisition was 5.9 (95 percent CI: 2.8, 12.5) for recent incident HSV-2, 3.1 (95 percent CI: 1.6, 6.1) for remote incident HSV-2, and 1.8 (95 percent CI: 1.3, 2.4) for prevalent HSV-2, compared with HSV-2 seronegative participants. When adjusted for the important markers of risk behavior as described in detail by Koblin et al. (21Go), the hazard ratio of HIV acquisition among those with recent incident HSV-2 was 3.6 (95 percent CI: 1.7, 7.8); that for remote incident HSV-2 was 1.7 (95 percent CI: 0.8, 3.3), and that for prevalent HSV-2 was 1.5 (95 percent CI: 1.1, 2.1) (table 2), compared with participants who were HSV-2 seronegative. These findings are somewhat sensitive to assumptions about the timing of infection for the 37 participants with uncertain acquisition dates. The adjusted hazard ratios for recent incident and remote incident HSV-2 vary from 2.6 (95 percent CI: 1.1, 6.1) to 6.3 (95 percent CI: 3.7, 10.9) and from 1.6 (95 percent CI: 0.8, 3.2) to 0.7 (95 percent CI: 0.2, 2.1), respectively, as the time of HSV-2 infection is varied from the beginning to the end of the follow-up period for these 37 participants.


Figure 2
View larger version (9K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		FIGURE 2. Human immunodeficiency virus (HIV) acquisition rates by herpes simplex virus type 2 (HSV-2) status among EXPLORE study participants in six US cities, 1999–2003. Bars represent 95% confidence intervals.

 

View this table:
[in this window]
[in a new window]

  		TABLE 2. Factors associated with human immunodeficiency virus acquisition among men who have sex with men in six US cities participating in the EXPLORE study, 1999–2003

 

   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
References
 
This multicenter trial is the largest prospective study of HSV-2 incidence and risk factors ever conducted among men who have sex with men, with 167 documented HSV-2 seroconversions, and the first study to estimate HSV-2 acquisition rates on a per-contact basis among men who have sex with men, although these estimates are limited because the HSV-2 serostatus of their partners is unknown. It is also the only study to ascertain the effect of a behavioral intervention on HSV-2 acquisition among men who have sex with men. The intensive, 10-session risk-reduction behavioral intervention tested in the EXPLORE study did not reduce HIV incidence to a statistically significant degree and had no effect on HSV-2 incidence, despite the statistically significant reduction in reported unprotected receptive anal intercourse (11Go). Participants with HSV-2 infection in this cohort had elevated rates of HIV acquisition; furthermore, these data suggest that incident HSV-2 might confer a further increase in HIV risk compared with prevalent HSV-2 infection.

The elevated risk of HIV infection among HSV-2 seropositive participants in this cohort substantiates the findings of meta-analyses by Wald and Link (7Go) and Freeman et al. (8Go). In the latter meta-analysis, the relative risk of HIV acquisition among individuals with HSV-2 infection ranged from 1.7 (95 percent CI: 1.2, 2.4) for men who have sex with men to 3.1 (95 percent CI: 1.7, 5.6) for women. Additionally, the stronger association between HIV and recently acquired HSV-2, compared with prevalent HSV-2, observed in this cohort of men who have sex with men is consistent with the temporal relation between HSV-2 and HIV acquisition that has been suggested by several prior studies among heterosexual men and women in India and Africa (4Go, 6Go, 24Go, 25Go). For example, in a cohort of sexually transmitted disease clinic patients in India, the hazard ratio of HIV was over twice as high among participants with incident HSV-2 infection (HSV-2 acquired within 6 months or less) compared with participants who were HSV-2 seropositive at enrollment (HR = 3.81 vs. 1.67) (4Go). In a nested case-control study of HIV seroconversions in Mwanza, Tanzania, the incidence of HIV infection was higher (odds ratios = 5.6 for men and 4.8 for women) among persons who seroconverted to HSV-2 during follow-up than among those who were HSV-2 seropositive at enrollment (odds ratios = 3.7 for men and 2.9 for women) compared with HSV-2 seronegative subjects (25Go).

In these studies and the present study, many participants seroconverted to both HSV-2 and HIV during the same follow-up interval; thus, the temporal sequence of the two infections can not be ascertained for all participants. The stronger association of incident HSV-2 with HIV infection is biologically plausible as the greater frequency, severity, and duration of reactivations of HSV-2 early after HSV-2 acquisition may increase the chance that HSV-2 is reactivating when a man is exposed to HIV. However, this association may also simply represent the simultaneous acquisition of HSV-2 and HIV from a dually infected partner. Given that up to 70 percent of HIV-positive men who have sex with men are also HSV-2 seropositive (26Go–28Go) and that rates of HSV-2 shedding are over three times higher in HIV-positive compared with HIV-negative individuals (29Go), exposure to both viruses during sexual contact with a high-risk partner is likely. Distinguishing between these two possibilities (a causal association vs. simultaneous acquisition) is very difficult and would require very frequent HSV-2 and HIV testing of study subjects. Another possibility is that some of the increased risk of HIV observed with incident HSV-2 infection may be due to residual confounding by factors that elevate risk of acquiring both HSV-2 and HIV. Although the hazard ratios for recent and remote incident HSV-2 were substantially attenuated by adjustment for sexual risk behaviors, there may be further confounding by inaccuracies in self-reported behavior or by unmeasured variables, such as genetic factors, that might influence risk of both HSV-2 and HIV acquisition.

The risk factors for HSV-2 acquisition in this cohort were unprotected receptive anal intercourse, a higher number of male sex partners, having an HIV-positive male partner, and African-American race. The increased risk of HSV-2 among African Americans in this cohort likely reflects the higher HSV-2 prevalence among African Americans (30Go, 31Go) combined with the tendency toward assortative mating (choosing a sexual partner of one's own racial/ethnic group) (32Go, 33Go).

This is the first study to estimate per-contact HSV-2 acquisition rates among men who have sex with men. Because the HSV-2 serostatus of participants' partners was unknown, we included all reported sexual acts in our per-contact analyses, recognizing that HSV-2 prevalence ranges from at least 20 percent in HIV-negative men who have sex with men to up to 70 percent among HIV-positive men who have sex with men (26Go, 28Go). The per-contact HSV-2 acquisition rates we have calculated are likely to be attenuated by the inclusion of not-at-risk acts (i.e., those with HSV-2 seronegative partners). If the prevalence of HSV-2 among the partners of participants is estimated to be 20 percent (the baseline prevalence in this cohort), then the true per-contact rates derived from at-risk acts might be expected to be approximately fivefold higher than the rates presented here. The increase in per-contact HSV-2 acquisition rates seen after exclusion of observations containing the highest numbers of sex acts indicates that the model assumptions of independence are not met, leading to further uncertainty in estimated rates. Our lack of information about the distribution of sex acts across partners and partner's HSV-2 status, coupled with the evidence from sensitivity analyses that our results were quite sensitive to model assumptions, indicates that these estimates of per-contact rates could be inaccurate and should be interpreted as rough estimates. However, it would be difficult to obtain better estimates of per-contact rates without conducting a study among couples of HSV-2 discordant monogamous men who have sex with men. The higher per-contact HSV-2 acquisition rates observed for protected acts compared with unprotected acts may indicate that participants modified their behavior on the basis of their perception of their partner's risk; participants may have been more likely to use a condom with a partner perceived to be high risk, as has been suggested by Warner et al. (34Go) and Casper et al. (35Go).

Strengths of this study include the high retention rate over up to 4 years of follow-up (87 percent in the control arm and 83 percent in the intervention arm) (11Go). The use of highly accurate serologic tests for HSV-2 and HIV allowed for precise outcome measurement. Collection of extensive behavioral data by use of audio computer-assisted self-interview allowed for a detailed assessment of HSV-2 risk factors and enabled adjustment for potential behavioral confounders of the relation between HSV-2 infection and HIV acquisition. Although the intervention did not reduce HIV or HSV-2 acquisition rates to a statistically significant degree, the control arm received extensive counseling at semiannual visits based on the best available model (Project RESPECT (14Go)), which may have decreased the chance of detecting a difference between study arms. The absence of a decrease in HSV-2 acquisition, despite the significant reduction in self-reported unprotected receptive anal intercourse with an HIV-positive or unknown serostatus partner, may reflect the importance of unmeasured factors related to partner selection and is consistent with prior observations that the efficacy of behavioral interventions may be underestimated by self-reported behavior change (11Go, 36Go).

This study had several limitations. Limited demographic data were available on the partners of participants. No herpes simplex virus type 1 (HSV-1) testing was done in this study, because of issues of cost and the need for oral and genital cultures to determine the site of HSV-1 infection. Since HSV-1 is an increasingly common cause of genital herpes, particularly among men who have sex with men (37Go), this study may have underestimated the impact of genital herpes (due to either HSV-1 or HSV-2) on the risk of HIV acquisition in this population, although when assessed serologically, HSV-1 has not been found to be a risk factor for HIV acquisition among men who have sex with men (5Go). The finding that 17 percent of HSV-2 seronegative participants did report genital ulcer disease symptoms indicates that genital HSV-1 infections may have been relatively common in this cohort, especially given the low reported rates of syphilis. However, syphilis data were collected by self-report (of syphilis diagnosed since prior study visit) rather than by serologic screening of all participants; thus, syphilis may have been underdiagnosed in this cohort.

The timing of HSV-2 acquisition had to be imputed for 27 HSV-2 seroconverters who had no interim sera available for HSV-2 testing and was assigned on the basis of lesions for an additional 10 seroconverters. Sensitivity analyses revealed that these assumptions may have influenced the hazard ratios we obtained for the associations of recent and remote incident HSV-2 with HIV acquisition. All participants who were HSV-2 seropositive at enrollment were classified as having prevalent HSV-2 at all study visits, with potential misclassification as some of these participants likely acquired HSV-2 recently before enrollment into the EXPLORE study, given the annual 1.9 percent HSV-2 incidence in this population. However, the bias produced by this misclassification is expected to be conservative. Finally, for over half (13 of 20) of the participants who acquired both HSV-2 and HIV during follow-up, the temporal sequence of the two infections could not be ascertained, because both were detected at the same study visit. This has been a limitation of all prior studies of the association of incident and prevalent HSV-2 with HIV acquisition.

Since this intensive behavioral intervention did not reduce HSV-2 acquisition, it is important that other methods of HSV-2 prevention be evaluated among men who have sex with men, including education, counseling about HSV-2 based on type-specific serologic testing, HSV-2 suppressive therapy, and candidate HSV-2 vaccines. Serologic testing is needed to identify men who have sex with men with HSV-2 infection, as symptoms were neither sensitive nor specific for prevalent or incident HSV-2 infection. It is evident from these data that HSV-2 infection is an important risk factor for HIV acquisition among men who have sex with men, although it remains uncertain whether incident HSV-2 confers additional risk compared with prevalent HSV-2. Even considering only the prevalent HSV-2 infections, at least 33 percent of HIV acquisitions among HSV-2-infected participants in this cohort are attributable to HSV-2, on the basis of the adjusted hazard ratio of 1.5. Given these findings, HSV-2 infection appears to be an important target for HIV prevention interventions. Studies of HSV-2 suppression with acyclovir to prevent HIV acquisition and transmission are currently underway (38Go) and will help to further clarify the role of HSV-2 in the spread of HIV.


   ACKNOWLEDGMENTS
 
This work was supported by the HIV Prevention Trials Network and sponsored by the US National Institute of Allergy and Infectious Diseases and the National Institute on Alcohol Abuse and Alcoholism of the National Institutes of Health, US Department of Health and Human Services, through contract N01 AI35176 with Abt Associates, Inc.; contract N01 AI45200 with the Fred Hutchinson Cancer Research Center; and subcontracts with the Denver Public Health, the Fenway Community Health Center, the Howard Brown Health Center, the New York Blood Center, the Public Health Foundation, Inc., and the University of Washington. In addition, this work was supported by the HIV Prevention Trials Network and sponsored by the National Institute of Allergy and Infectious Diseases, the National Institute of Child Health and Human Development, the National Institute on Drug Abuse, the National Institute of Mental Health, and the Office of AIDS Research of the National Institutes of Health, US Department of Health and Human Services, through a cooperative agreement with Family Health International (cooperative agreement 5 U01 AI46749) with a subsequent subcontract to Abt Associates, Inc., with subcontracts to the Howard Brown Health Center and Denver Public Health; cooperative agreement U01 AI48040 to the Fenway Community Health Center; cooperative agreement U01 AI48016 to Columbia University (including a subagreement with the New York Blood Center); cooperative agreement U01 AI479781 to the University of Washington; and cooperative agreement U01 AI47995 to the University of California, San Francisco. E. L. B. was supported by National Institutes of Health training grant T32 AI-07140. A. W. was supported by National Institute of Allergy and Infectious Diseases grant AI-30731.

The authors thank Geetha Beauchamp, Anne Cent, Dr. Margaret Chesney, Dr. Thomas Coates, Dr. Deborah Donnell, and Brent Sugimoto for their contributions to this study, as well as the dedicated Project EXPLORE site staff.

Dr. Anna Wald receives grant support from GlaxoSmithKline, Antigenics, Roche, and Vical. She is a consultant for Novartis, Powdermed, and Medigene.


   References
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 References
 

  1. Kaiser Family Foundation. Sexually transmitted diseases in America: how many cases and at what cost? Research Triangle Park, NC: American Social Health Association, 1998.
  2. Malkin JE. Epidemiology of genital herpes simplex virus infection in developed countries. Herpes 2004;11(suppl 1):2A–23A.
  3. Weiss H. Epidemiology of herpes simplex virus type 2 infection in the developing world. Herpes 2004;11(suppl 1):24A–35A.
  4. Reynolds SJ, Risbud AR, Shepherd ME, et al. Recent herpes simplex virus type 2 infection and the risk of human immunodeficiency virus type 1 acquisition in India. J Infect Dis 2003;187:1513–21.[CrossRef][ISI][Medline]
  5. Renzi C, Douglas JM Jr, Foster M, et al. Herpes simplex virus type 2 infection as a risk factor for human immunodeficiency virus acquisition in men who have sex with men. J Infect Dis 2003;187:19–25.[CrossRef][ISI][Medline]
  6. del Mar Pujades Rodriguez M, Obasi A, Mosha F, et al. Herpes simplex virus type 2 infection increases HIV incidence: a prospective study in rural Tanzania. AIDS 2002;16:451–62.[CrossRef][ISI][Medline]
  7. Wald A, Link K. Risk of human immunodeficiency virus (HIV) infection in herpes simplex virus type-2 (HSV-2) seropositive persons: a meta-analysis. J Infect Dis 2002;185:45–52.[CrossRef][ISI][Medline]
  8. Freeman EE, Weiss HA, Glynn JR, et al. Herpes simplex virus 2 infection increases HIV acquisition in men and women: systematic review and meta-analysis of longitudinal studies. AIDS 2006;20:73–83.[ISI][Medline]
  9. Koelle DM, Abbo H, Peck A, et al. Direct recovery of herpes simplex virus (HSV)-specific T lymphocyte clones from recurrent genital HSV-2 lesions. J Infect Dis 1994;169:956–61.[ISI][Medline]
  10. Koelle DM, Benedetti J, Langenberg A, et al. Asymptomatic reactivation of herpes simplex virus in women after first episode of genital herpes. Ann Intern Med 1992;116:433–7.[ISI][Medline]
  11. Koblin B, Chesney M, Coates T. Effects of a behavioural intervention to reduce acquisition of HIV infection among men who have sex with men: the EXPLORE randomised controlled study. Lancet 2004;364:41–50.[CrossRef][ISI][Medline]
  12. Chesney MA, Koblin BA, Barresi PJ, et al. An individually tailored intervention for HIV prevention: baseline data from the EXPLORE study. Am J Public Health 2003;93:933–8.[Abstract/Free Full Text]
  13. Koblin BA, Chesney MA, Husnik MJ, et al. High-risk behaviors among men who have sex with men in 6 US cities: baseline data from the EXPLORE study. Am J Public Health 2003;93:926–32.[Abstract/Free Full Text]
  14. Kamb ML, Fishbein M, Douglas JM Jr, et al. Efficacy of risk-reduction counseling to prevent human immunodeficiency virus and sexually transmitted diseases: a randomized controlled trial. Project RESPECT Study Group. JAMA 1998;280:1161–7.[Abstract/Free Full Text]
  15. Hewett PC, Mensch BS, Erulkar AS. Consistency in the reporting of sexual behaviour by adolescent girls in Kenya: a comparison of interviewing methods. Sex Transm Infect 2004;80(suppl 2):ii43–8.[Abstract/Free Full Text]
  16. Kurth AE, Martin DP, Golden MR, et al. A comparison between audio computer-assisted self-interviews and clinician interviews for obtaining the sexual history. Sex Transm Dis 2004;31:719–26.[ISI][Medline]
  17. Metzger DS, Koblin BA, Turner C, et al. Randomized controlled trial of audio computer-assisted self-interviewing: utility and acceptability in longitudinal studies. HIVNET Vaccine Preparedness Study Protocol Team. Am J Epidemiol 2000;152:99–106.[Abstract/Free Full Text]
  18. Ashley RL, Militoni J, Lee F, et al. Comparison of Western blot (immunoblot) and glycoprotein G-specific immunodot enzyme assay for detecting antibodies to herpes simplex virus types 1 and 2 in human sera. J Clin Microbiol 1988;26:662–7.[Abstract/Free Full Text]
  19. Morrow RA, Friedrich D, Meier A, et al. Use of "biokit HSV-2 Rapid Assay" to improve the positive predictive value of Focus HerpeSelect HSV-2 ELISA. BMC Infect Dis 2005;5:84. (doi:10.1186/1471-2334-5-84).[CrossRef][Medline]
  20. Golden MR, Ashley-Morrow R, Swenson P, et al. Herpes simplex virus type 2 (HSV-2) Western blot confirmatory testing among men testing positive for HSV-2 using the focus enzyme-linked immunosorbent assay in a sexually transmitted disease clinic. Sex Transm Dis 2005;32:771–7.[CrossRef][ISI][Medline]
  21. Koblin BA, Husnik MJ, Colfax G, et al. Risk factors for HIV infection among men who have sex with men. AIDS 2006;20:731–9.[ISI][Medline]
  22. Mirowsky J, Ross CE. Age and depression. J Health Soc Behav 1992;33:187–205; discussion 206–12.[CrossRef][ISI][Medline]
  23. Wacholder S. Binomial regression in GLIM: estimating risk ratios and risk differences. Am J Epidemiol 1986;123:174–84.[Abstract/Free Full Text]
  24. McFarland W, Gwanzura L, Bassett MT, et al. Prevalence and incidence of herpes simplex virus type 2 infection among male Zimbabwean factory workers. J Infect Dis 1999;180:1459–65.[CrossRef][ISI][Medline]
  25. Todd J, Grosskurth H, Changalucha J, et al. Risk factors influencing HIV infection incidence in a rural African population: a nested case-control study. J Infect Dis 2006;193:458–66.[CrossRef][ISI][Medline]
  26. Meyer JL, Crosby RA, Whittington WL, et al. The psychosocial impact of serological herpes simplex type 2 testing in an urban HIV clinic. Sex Transm Infect 2005;81:309–15.[Abstract/Free Full Text]
  27. Stamm WE, Handsfield HH, Rompalo AM, et al. Association between genital ulcer disease and acquisition of HIV infection in homosexual men. JAMA 1988;260:1429–33.[Abstract]
  28. Russell DB, Tabrizi SN, Russell JM, et al. Seroprevalence of herpes simplex virus types 1 and 2 in HIV-infected and uninfected homosexual men in a primary care setting. J Clin Virol 2001;22:305–13.[CrossRef][ISI][Medline]
  29. Schacker T, Zeh J, Hu HL, et al. Frequency of symptomatic and asymptomatic HSV-2 reactivations among HIV-infected men. J Infect Dis 1998;178:1616–22.[CrossRef][ISI][Medline]
  30. Fleming D, McQuillan G, Johnson R, et al. Herpes simplex virus type 2 in the United States, 1976 to 1994. N Engl J Med 1997;337:1105–11.[Abstract/Free Full Text]
  31. Turner KR, McFarland W, Kellogg TA, et al. Incidence and prevalence of herpes simplex virus type 2 infection in persons seeking repeat HIV counseling and testing. Sex Transm Dis 2003;30:331–4.[ISI][Medline]
  32. Laumann EO, Youm Y. Racial/ethnic group differences in the prevalence of sexually transmitted diseases in the United States: a network explanation. Sex Transm Dis 1999;26:250–61.[ISI][Medline]
  33. Mark HD, Sifakis F, Hylton JB, et al. Sex with women as a risk factor for herpes simplex virus type 2 among young men who have sex with men in Baltimore. Sex Transm Dis 2005;32:691–5.[CrossRef][ISI][Medline]
  34. Warner L, Newman DR, Austin HD, et al. Condom effectiveness for reducing transmission of gonorrhea and chlamydia: the importance of assessing partner infection status. Am J Epidemiol 2004;159:242–51.[Abstract/Free Full Text]
  35. Casper C, Carrell D, Miller KG, et al. HIV serodiscordant sex partners and the prevalence of human herpesvirus 8 infection among HIV negative men who have sex with men: baseline data from the EXPLORE study. Sex Transm Infect 2006;82:229–35.[Abstract/Free Full Text]
  36. Imrie J, Stephenson JM, Cowan FM, et al. A cognitive behavioural intervention to reduce sexually transmitted infections among gay men: randomised trial. BMJ 2001;322:1451–6.[Abstract/Free Full Text]
  37. Lafferty WE, Downey L, Celum C, et al. Herpes simplex virus type 1 as a cause of genital herpes: impact on surveillance and prevention. J Infect Dis 2000;181:1454–7.[CrossRef][ISI][Medline]
  38. Celum CL, Robinson NJ, Cohen MS. Potential effect of HIV type 1 antiretroviral and herpes simplex virus type 2 antiviral therapy on transmission and acquisition of HIV type 1 infection. J Infect Dis 2005;191(suppl 1):S107–14.

Add to CiteULike CiteULike   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us    What's this?



This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow All Versions of this Article:
164/8/733    most recent
kwj270v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (7)
Right arrowRequest Permissions
Right arrow Disclaimer
Google Scholar
Right arrow Articles by Brown, E. L.
Right arrow Articles by Celum, C.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Brown, E. L.
Right arrow Articles by Celum, C.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?