Human herpesvirus-8 (HHV-8) is frequently detected in oropharyngeal secretions from HIV-infected men who have sex with men (MSM), and contact with saliva may be an important mode of HHV-8 transmission. A total of 196 MSM were followed over 2 years to determine the correlates of HHV-8 oropharyngeal shedding. A total of 134 (68%) of 196 participants were HHV-8 seropositive upon enrollment, and 9 (15%) of 62 participants seroconverted to HHV-8 during follow-up. HHV-8 DNA was detected in 43 (22%) of 196 participants: 39 (27%) of 134 HHV-8 seropositive, 4 (8%) of 53 HHV-8 seronegative, and 5 (56%) of 9 seroconverters to HHV-8. HHV-8 was detected in 101 (15%) of 696 total oral specimens: 84 (17%) of 481 samples from HHV-8–seropositive men, 6 (3%) of 180 samples from HHV-8–seronegative men, and 11 (31%) of 35 samples from seroconverters. Using adjusted marginal structural models, HHV-8 shedding was higher in men not receiving highly active antiretroviral therapy (odds ratio 2.4, 95% CI 1.0–6.0, P = 0.06), with CD4 counts > 200 cells/mm3 (odds ratio 4.8, 95% CI 1.0–22.8, P = 0.05), or with detectable oral leukocyte esterase (odds ratio 5.0, 95% CI 2.0–12.5, P < 0.01). CD4 count, antiretroviral therapy, and oral inflammation may influence HHV-8 oropharyngeal shedding.
Infection with human herpesvirus-8 (HHV-8) is common among men who have sex with men (MSM). 1–8 Although no specific behavior has consistently been associated with acquisition of HHV-8 infection, HHV-8 may be transmitted through saliva. 9 HHV-8 DNA quantity is highest in saliva; oral epithelial cells are competent to act as a source of replication for the virus; and deep kissing has been associated with HHV-8 infection. 9 In a cohort of prospectively followed MSM, a dichotomous pattern of HHV-8 shedding was noted, with some men having HHV-8 detectable in the saliva on a majority of days and others failing to demonstrate detectable virus during the 15- to 75-day collection period. If oral contact is an important mode of transmission for HHV-8, understanding the determinants of oropharyngeal viral shedding may have important ramifications for prevention of disease. Therefore, we conducted a study to evaluate the predictors of HHV-8 shedding among HIV-seropositive MSM.
From the *Department of Medicine, The University of Washington, Seattle, WA; †Department of Biostatistics, The University of Washington, Seattle, WA; ¶Department of Epidemiology, The University of Washington, Seattle, WA; ††Department of Pathology, The University of Washington, Seattle, WA; **Department of Laboratory Medicine, The University of Washington, Seattle, WA; ‡Division of Infectious Disease, Fred Hutchinson Cancer Research Center, Seattle, WA; and §The Polyclinic, Seattle, WA.
Received for publication January 28, 2003; accepted November 26, 2003.
Support: NIH Grants 5R01CA055488-09, P01 AI30731 and U19 AI31448 and the Joel Meyer Infectious Disease Scholarship Grant.
Reprints: Corey Casper, University of Washington, Virology Research Clinic, 600 Broadway, Suite 400, Seattle, WA 98122 (e-mail: email@example.com).
MATERIALS AND METHODS
Between July 1996 and December 1998, 275 HIV-infected MSM were enrolled into a study of HIV-1 and the development of anal dysplasia as previously described. 10 Men in Seattle, WA, were recruited from outpatient clinics and advertisements in the community. Written informed consent was obtained in accordance with a protocol approved by the University of Washington Human Subjects Division.
Participants were seen from 1–6 visits over a period ranging from 6 months–3 years. At each study visit, a self-administered questionnaire was used to obtain information regarding demographic characteristics, sexual behavior, and medical histories (including any medications taken during the week prior to the study visit). Specific questions were asked regarding the duration of HIV infection, number of sexual partners, and the history of sexually transmitted infections (STIs). General physical examinations were also performed.
Specimen Collection and Laboratory Testing
Serum and blood anticoagulated with ethylenediamine tetra-acetic acid (EDTA) were collected at each study visit. Oral secretions were collected at each visit with 10 mL of sterile saline and frozen for DNA extraction as previously described. 11
Serologic assays for HHV-8 were conducted using an immunofluorescence assay (IFA) with HHV-8-positive BCBL-1 lymphoma cells. 12 Seropositivity was defined as the presence of lytic antibodies, with or without latency-associated nuclear antigen at a serum dilution of at least 1:40. All were reacted against a BJAB cell line to detect nonspecific reactivity. The first and last serum samples from each participant were tested for antibodies to HHV-8, along with all available interval specimens from men who seroconverted.
HHV-8 DNA was measured quantitatively using a fluorescent probe-based polymerase chain reaction (PCR) assay (TaqMan assay, Applied Biosystems, Foster City, CA) as described previously. 9,13–15 Oral specimens were subjected to DNA extraction, purification, and amplification with primers to the orf73 gene (ORF73-F: 5`-CCA GGA AGT CCC ACA GTG TTC-3`, ORF73-R: 5`-GCC ACC GGT AAA GTA GGA CTA GAC-3`). All samples were run with positive and negative controls. Six specimens were selected at random to assess the uniqueness of each HHV-8 DNA amplimer using heteroduplex mobility assay (HMA) to a 672-basepair section of the K1 hypervariable region. 16 Two additional specimens, representing HHV-8 DNA amplified from the same participant at 2 visits separated by 3 months in time, was also subjected to HMA to assess the intraperson variability of detected strains. Primer set K1-A,B (TGC CCT GGA GTG ATT TCA AC and AGA TGC CAA ACG GTA ACA TTA TTT C) and K1-C,D (ATA ATG TTA CCG TTT GGC ATC TAC C and TGG CAC TGT TTT GTT TGA GTC A) amplified 305- and 416-basepair fragments, respectively, as previously described. 14 PCR cycle conditions were modified to an initial cycle of 96°C for 3 minutes followed by 50 cycles of 96°C (30 seconds), 54°C (30 seconds), and 72°C (30 seconds) and ending with a single cycle of 72°C for 5 minutes. PCR products K1-A,B and K1-C,D were analyzed by heteroduplex formation as described previously. 17 To generate probe, 20 ng QIAEX II–purified BCBL-1 PCR products were end-labeled with P32 by T4 polynucleotide kinase in a final volume of 20 μL. Four to 6 μL of each nonpurified PCR reaction was hybridized to 0.5 μL of probe. Each hybridization reaction consisted of 0.5 μL of probe, 4–6 μL of PCR reaction, 200 mM NaCl, 10 mM tromethamine hydrochloride (pH 8.0), and 2 mM EDTA. After denaturation at 95°C for 5 minutes and annealing at 55°C for 2 hours, the entire hybridization reaction was loaded on 1-mm-thick mutation detection enhancement (MDE) gel (FMC BioProducts, Rockland, ME) and electrophoresed for 20 hours at 500 V. The gel was dried and exposed to x-ray film. The hybridization reactions of BCBL-1 PCR reactions to the probes were used as the marker of homoduplex formation.
HIV-1 RNA was measured using the AMPLICOR Monitor HIV-1 Test (Roche, Alameda, CA). Oral inflammation was assessed semiquantitatively with the use of leukocyte esterase chemical indicator strips (Bayer, Elkhart, IN).
Infection with HHV-8 was defined as either a positive serologic assay or the presence of HHV-8 DNA detected from saliva by PCR. Highly active antiretroviral therapy (HAART) was defined as the use of ≥3 antiretroviral agents with at least 1 protease inhibitor or nonnucleoside reverse transcriptase inhibitor at the time of the study visit. Differences between HHV-8–seropositive and seronegative MSM were assessed using χ2 tests (dichotomous variables), Mantel-Haenszel test for trend (ordered categorical variables), or t tests (continuous variables).
The analysis of HHV-8 shedding over time included all samples from seropositive subjects. Because HAART use is indicated by factors such as low CD4 count or high plasma HIV-1 RNA level, the observed association between HAART and HHV-8 shedding may be confounded by the indication for HAART use. To address this, we used inverse probability of treatment-weighted (IPTW) estimation and generalized estimating equations (PROC GENMOD, SAS Institute, Cary, NC) with a logit link to obtain parameter estimates and robust estimates of the variance. 18,19 IPTW uses the predicted probability of actual HAART received (or lack of), conditional on the potential confounders of CD4 count and plasma HIV-1 RNA level to increase or decrease the weight given to visits that might be under- (i.e., HAART use in the setting of high CD4 counts) or overrepresented (i.e., HAART use among men with low CD4 counts or high plasma HIV-1 RNA level) due to confounding. A 2-sided 0.05 level test determined statistical significance for all analyses.
Baseline Study Characteristics
A total of 196 (71%) of 275 participants had specimens suitable for this analysis. At enrollment, 134 (68%) of 196 participants were HHV-8 seropositive (Table 1). HHV-8–seropositive persons were more likely to have used antiviral medications in the past (P = 0.003), were HIV-infected for a greater number of years (mean 9.8 vs. 8.4 years, P = 0.05), and had more advanced HIV disease with lower CD4 counts (mean 386 vs. 483 cells/mm3, P = 0.02). No significant differences were found in the number of sexual partners (P = 0.83) or history of STI (P = 0.20) between HHV-8–seronegative and –seropositive persons. An additional 9 (15%) of 62 participants seroconverted to HHV-8 during the course of follow-up.
Frequency of HHV-8 Shedding
HHV-8 DNA was detected in oral specimens from 43 (22%) of 196 participants, including 39 (27%) of 134 HHV-8– seropositive, 4 (8%) of 53 persistently seronegative subjects, and 5 (56%) of 9 men who seroconverted to HHV-8 (P = 0.06 for seropositive vs. seronegative, P = 0.003 for seroconverters vs. seronegative, P = 0.002 for seroconverters vs. seropositive). Of men with detectable HHV-8 DNA, 10 (23%) had HHV-8 detected at 1 visit, 18 (42%) at 2 visits, 8 (19%) at 3 visits, 5 (12%) at 4 visits, 1 (2%) at 5 visits, and 1 (2%) at 6 visits. Overall, HHV-8 was detected in 101 (15%) of 696 oral specimens including 84 (17%) of 481 samples from seropositive men, 11 (31%) of 35 specimens from men who seroconverted, and 6 (3%) of 180 specimens from seronegative men.
Correlates of HHV-8 Shedding
HHV-8 DNA was recovered more frequently from the oropharynx at visits when an individual was not receiving HAART (25 vs. 16%, odds ratio [OR] 2.4, 95% CI 1.0–6.0, P = 0.06), had a CD4 count of >200 cells/mm3 (20 vs. 5%, OR 4.8, 95% CI 1.0–22.8, P = 0.05), or had detectable leukocyte esterase in the oropharynx (21 vs. 7%, OR 5.0, 95% CI 2.0–12.5, P < 0.01) (Table 2). No significant association was found between HHV-8 shedding and duration of HIV infection (P = 0.39), age (P = 0.41), number of sexual partners in the past year (P = 0.07), or HIV-1 plasma RNA level (P = 0.63).
Selected HHV-8 Strain Typing
All participants selected for HMA analysis had unique HHV-8 strains amplified from their oropharynx (Fig. 1). Neither of the 2 randomly selected PCR-negative samples had detectable HMA bands, and the 2 samples obtained from the same patient over a 3-month period had identical HMA banding patterns.
HHV-8 seroprevalence was high among this cohort of HIV-positive MSM, a group that has the highest prevalence of HHV-8 infection in North America and Europe. Approximately 50% of HIV-infected MSM without Kaposi sarcoma (KS) were determined to have serum antibodies to HHV-8 by IFA in 2 previous studies. 20,21 In our study, 68% of participants were HHV-8 seropositive upon enrollment, a fact that may be accounted for by the high proportion of men reporting multiple sexual partners or a history of STIs. The proportion of HHV-8–seropositive MSM with HHV-8 detectable in the oropharynx was similar to that seen in a previous study, 9 but our study is the first to note the nearly twice as high a rate of oropharyngeal shedding among HHV-8 seroconverters as compared with persons chronically infected with HHV-8. Detection of HHV-8 DNA in the oropharynx in most men with serologic evidence of recent HHV-8 infection supports the early colonization of this site in the course of HHV-8 infection. The observation of 4 persistently seronegative participants with HHV-8 DNA detected in the oropharynx suggests that development or detection of antibody may not be universal in infected persons, or that seroconversion may lag by many months after acquisition. 22 The use of stringent PCR methods, the failure to detect HHV-8 DNA in any negative controls, and the HMA results argue against false-positive PCR reactions. Two other studies have documented negative HHV-8 serologic tests in the setting of detectable HHV-8 DNA in the blood 23 and saliva. 24
We found that lack of HAART use and higher CD4 counts are correlated independently with increased frequency of HHV-8 shedding, but found no association with HIV-1 plasma viral load or other antiviral medication use. Since the introduction of HAART, a dramatic decline in the number of cases of KS has been observed. 25 The lower incidence rates of KS in the United States since 1996 do not seem to be explained by a drop in HHV-8 seroprevalence, 26 and the immune reconstitution afforded by potent antiretroviral therapy has been offered as one explanation for the decline in KS. A recent report found that HHV-8 viremia was significantly lower after the initiation of HAART among MSM with KS. 27 HAART may influence HHV-8 infection among HIV-positive persons in a number of ways: the action of Tat or Vpr, 28,29 changes in the cytokine milieu, 30,31 or the exertion of a direct antiviral effect on HHV-8 by ≥1 HAART medicines. 32–34
HHV-8 shedding was more frequent among men with higher CD4 counts, which is similar to what has been described among a cohort of HIV-positive women, 35 and is consistent with studies of the other human gamma-herpesvirus, Epstein-Barr virus, 36,37 and the beta-herpesvirus, human herpesvirus-6. 38 It is possible that persons with increased CD4 counts may have increased amounts of inflammatory cytokines, which are permissive for HHV-8 replication. 30
We also found an association between inflammation in the oral cavity and HHV-8 shedding, a relationship that has not been previously described. No relationship between HHV-8 vaginal or cervical shedding and clinical evidence of genital inflammation or ulceration was found in a small number of Zimbabwean women with KS. 11 However, HIV-1 is shed with higher frequency from urethral secretions in persons with urethritis, 39 and HIV-1 shedding in cervical secretions has been strongly correlated with tumor necrosis factor-α and interleukin-6 concentrations. 40 It is unclear from the current study whether inflammation is a consequence of HHV-8 infection or if HHV-8 shedding is more common from inflamed oral mucosa; future research should be aimed at determining the temporal relationship between these 2 factors.
This project was not initially designed to study HHV-8 infection and is limited by the infrequent sampling of oropharyngeal secretions. We have demonstrated in studies of other herpesvirus infections that daily sampling at mucosal sites is required to provide an accurate picture of the burden of viral replication in MSM. 41,42 Future studies with daily sampling will be important to more completely characterize the relationship between HIV-1 and HHV-8 oropharyngeal shedding.
If mucosal shedding frequency or quantity is important in determining the risk of acquiring HHV-8 from an infected sexual partner, then our data imply that the use of HAART therapy could decrease HHV-8 transmission. Recent HIV treatment guidelines suggest deferring HAART therapy until a substantial decrease in CD4 count. 43 Additional studies are needed to determine the effect of these recommendations on HHV-8 incidence rates in the future.
We are indebted to Stacy Selke for her careful review of laboratory data and analysis files. Support: NIH Grants 5R01CA055488-09, P01 AI30731 and U19 AI31448 and the Joel Meyer Infectious Disease Scholarship Grant.
1. Gao SJ, Kingsley L, Li M, et al. KSHV antibodies among Americans, Italians and Ugandans with and without Kaposi's sarcoma. Nat Med. 1996; 2:925–928.
2. Kedes DH, Operskalski E, Busch M, et al. The seroepidemiology of human herpesvirus 8 (Kaposi's sarcoma-associated herpesvirus): distribution of infection in KS risk groups and evidence for sexual transmission [published correction appears in Nat Med.
1996;2:1041]. Nat Med. 1996; 2:918–924.
3. Rainbow L, Platt GM, Simpson GR, et al. The 222- to 234-kilodalton latent nuclear protein (LNA) of Kaposi's sarcoma-associated herpesvirus (human herpesvirus 8) is encoded by orf73 and is a component of the latency-associated nuclear antigen. J Virol. 1997; 71:5915–5921.
4. Lennette ET, Blackbourn DJ, Levy JA. Antibodies to human herpesvirus type 8 in the general population and in Kaposi's sarcoma patients. Lancet. 1996; 348:858–861.
5. Simpson GR, Schulz TF, Whitby D, et al. Prevalence of Kaposi's sarcoma associated herpesvirus infection measured by antibodies to recombinant capsid protein and latent immunofluorescence antigen. Lancet. 1996; 348:1133–1138.
6. Melbye M, Cook PM, Hjalgrim H, et al. Risk factors for Kaposi's-sarcoma-associated herpesvirus (KSHV/HHV-8) seropositivity in a cohort of homosexual men, 1981–1996. Int J Cancer. 1998; 77:543–548.
7. Casper C, Wald A, Pauk J, et al. Correlates of prevalent and incident Kaposi's sarcoma-associated herpesvirus infection in men who have sex with men. J Infect Dis. 2002; 185:990–993.
8. Calabro ML, Sheldon J, Favero A, et al. Seroprevalence of Kaposi's sarcoma-associated herpesvirus/human herpesvirus 8 in several regions of Italy. J Hum Virol. 1998; 1:207–213.
9. Pauk J, Huang ML, Brodie SJ, et al. Mucosal shedding of human herpesvirus 8 in men. N Engl J Med. 2000; 343:1369–1377.
10. Lampinen TM, Critchlow CW, Kuypers JM, et al. Association of antiretroviral therapy with detection of HIV-1 RNA and DNA in the anorectal mucosa of homosexual men. AIDS. 2000; 14:F69–F75.
11. Lampinen TM, Kulasingam S, Min J, et al. Detection of Kaposi's sarcoma-associated herpesvirus in oral and genital secretions of Zimbabwean women. J Infect Dis. 2000; 181:1785–1790.
12. Chandran B, Smith MS, Koelle DM, et al. Reactivities of human sera with human herpesvirus-8-infected BCBL-1 cells and identification of HHV-8-specific proteins and glycoproteins and the encoding cDNAs. Virology. 1998; 243:208–217.
13. Heid CA, Stevens J, Livak KJ, et al. Real time quantitative PCR. Genome Res. 1996; 6:986–994.
14. Ryncarz AJ, Goddard J, Wald A, et al. Development of a high-throughput quantitative assay for detecting herpes simplex virus DNA in clinical samples. J Clin Microbiol. 1999; 37:1941–1947.
15. Zerr DM, Huang ML, Corey L, et al. Sensitive method for detection of human herpesviruses 6 and 7 in saliva collected in field studies. J Clin Microbiol. 2000; 38:1981–1983.
16. Lagunoff M, Ganem D. The structure and coding organization of the genomic termini of Kaposi's sarcoma-associated herpesvirus. Virology. 1997; 236:147–154.
17. Gretch DR, Polyak SJ, Wilson JJ, et al. Tracking hepatitis C virus quasispecies major and minor variants in symptomatic and asymptomatic liver transplant recipients. J Virol. 1996; 70:7622–7631.
18. Hernan MA, Brumback BA, Robins JM. Estimating the causal effect of zidovudine on CD4 count with a marginal structural model for repeated measures. Stat Med. 2002; 21:1689–1709.
19. Robins JM, Hernan MA, Brumback B. Marginal structural models and causal inference in epidemiology. Epidemiology. 2000; 11:550–560.
20. Verbeek W, Frankel M, Miles S, et al. Seroprevalence of HHV-8 antibodies in HIV-positive homosexual men without Kaposi's sarcoma and their clinical follow-up. Am J Clin Pathol. 1998; 109:778–783.
21. Smith NA, Sabin CA, Gopal R, et al. Serologic evidence of human herpesvirus 8 transmission by homosexual but not heterosexual sex. J Infect Dis. 1999; 180:600–606.
22. Goudsmit J, Renwick N, Dukers NH, et al. Human herpesvirus 8 infections in the Amsterdam Cohort Studies (1984–1997): analysis of seroconversions to ORF65 and ORF73. Proc Natl Acad Sci U S A. 2000; 97:4838–4843.
23. Wang QJ, Jenkins FJ, Jacobson LP, et al. Primary human herpesvirus 8 infection generates a broadly specific CD8(+) T-cell response to viral lytic cycle proteins. Blood. 2001; 97:2366–2373.
24. Casper C, Krantz E, Taylor H, et al. Assessment of a combined testing strategy for detection of antibodies to human herpesvirus 8 (HHV-8) in persons with Kaposi's sarcoma, persons with asymptomatic HHV-8 infection, and persons at low risk for HHV-8 infection. J Clin Microbiol. 2002; 40:3822–3825.
25. Jones JL, Hanson DL, Dworkin MS, et al. Effect of antiretroviral therapy on recent trends in selected cancers among HIV-infected persons. Adult/Adolescent Spectrum of HIV Disease Project Group. J Acquir Immune Defic Syndr. 1999; 21(Suppl 1):S11–S17.
26. Osmond DH, Buchbindes S, Cheng A, et al. Prevalence of Kaposi sarcoma-associated herpesvirus infection in homosexual men at beginning of and during the HIV epidemic. JAMA. 2002; 287:221–225.
27. Gill J, Bourboulia D, Wilkinson J, et al. Prospective study of the effects of antiretroviral therapy on Kaposi sarcoma-associated herpesvirus infection in patients with and without Kaposi sarcoma. J Acquir Immune Defic Syndr. 2002; 31:384–390.
28. Huang LM, Chao MF, Chen MY, et al. Reciprocal regulatory interaction between human herpesvirus 8 and human immunodeficiency virus type 1. J Biol Chem. 2001; 276:13427–13432.
29. Ensoli B, Gendelman R, Markham P, et al. Synergy between basic fibroblast growth factor and HIV-1 Tat protein in induction of Kaposi's sarcoma. Nature. 1994; 371:674–680.
30. Chang J, Renne R, Dittmer D, et al. Inflammatory cytokines and the reactivation of Kaposi's sarcoma-associated herpesvirus lytic replication. Virology. 2000; 266:17–25.
31. Ensoli B, Sturzl M, Monini P. Cytokine-mediated growth promotion of Kaposi's sarcoma and primary effusion lymphoma. Semin Cancer Biol. 2000; 10:367–381.
32. Gustafson EA, Schinazi RF, Fingeroth JD. Human herpesvirus 8 open reading frame 21 is a thymidine and thymidylate kinase of narrow substrate specificity that efficiently phosphorylates zidovudine but not ganciclovir. J Virol. 2000; 74:684–692.
33. Lock MJ, Thorley N, Teo J, et al. Azidodeoxythymidine and didehydrodeoxythymidine as inhibitors and substrates of the human herpesvirus 8 thymidine kinase. J Antimicrob Chemother. 2002; 49:359–366.
34. Leao JC, Kumar N, McLean KA, et al. Effect of human immunodeficiency virus-1 protease inhibitors on the clearance of human herpesvirus 8 from blood of human immunodeficiency virus-1-infected patients. J Med Virol. 2000; 62:416–420.
35. Ghandi M, Ameli N, Bacchetti P, et al. Salivary HHV8 shedding is more prevalent at higher CD4 counts. Paper presented at: 40th Annual Meeting of the Infectious Disease Society of America; October 2002; Chicago, IL.
36. Lowhagen GB, Bergbrant IM, Bergstrom T, et al. PCR detection of Epstein-Barr virus, herpes simplex virus and human papillomavirus from the anal mucosa in HIV-seropositive and HIV-seronegative homosexual men. Int J STD AIDS. 1999; 10:615–618.
37. Ferbas J, Rahman MA, Kingsley LA, et al. Frequent oropharyngeal shedding of Epstein-Barr virus in homosexual men during early HIV infection. Aids. Nov. 1992; 6:1273–1278.
38. Fairfax MR, Schacker T, Cone RW, et al. Human herpesvirus 6 DNA in blood cells of human immunodeficiency virus-infected men: correlation of high levels with high CD4 cell counts. J Infect Dis. 1994; 169:1342–1345.
39. Moss GB, Overbaugh J, Welch M, et al. Human immunodeficiency virus DNA in urethral secretions in men: association with gonococcal urethritis and CD4 cell depletion. J Infect Dis. 1995; 172:1469–1474.
40. Lawn SD, Subbarao S, Wright Jr, TC et al. Correlation between human immunodeficiency virus type 1 RNA levels in the female genital tract and immune activation associated with ulceration of the cervix. J Infect Dis. 2000; 181:1950–1956.
41. Wald A, Zeh J, Selke S, et al. Virologic characteristics of subclinical and symptomatic genital herpes infections. N Engl J Med. 1995; 333:770–775.
42. Krone MR, Wald A, Tabet SR, et al. Herpes simplex virus type 2 shedding in human immunodeficiency virus-negative men who have sex with men: frequency, patterns, and risk factors. Clin Infect Dis. 2000; 30:261–267.