Subjects and Methods

Study populationThe study was conducted at the Jefferson County Department of Health STD Clinic in Birmingham, Alabama. A sample of men presenting to the STD clinic for acute care who were evaluated by 2 providers (A.H. and J.M.S.) were approached for study enrollment. Men who identified themselves as MSMs were excluded, as were patients returning for follow-up of continuing problems. Most participants were attending the clinic for perceived STD symptoms (most commonly urethral discharge or dysuria), as partners of a person with a previously diagnosed STD, or for STD screening

Study proceduresHuman experimentation guidelines of the US Department of Health and Human Services and the University of Alabama at Birmingham were followed in the conduct of this study. After being given a description of study goals and procedures, patients were asked to provide written, informed consent. Consenting participants were then administered a standardized questionnaire collecting demographic, STD history, and sex partner data

After completing the questionnaire, participants underwent a directed physical exam. A serum sample was obtained for VDRL, voluntary HIV antibody testing (using a separate clinic consent), and HSV antibody testing. An initial voided urine sample was collected for nucleic acid amplification testing (Aptima Combo II; Gen-Probe), in accordance with the manufacturer’s instructions. Individual swab specimens from the urethra, shaft, coronal sulcus, glans, and foreskin (or foreskin remnant) were collected for herpes culture and PCR testing. If a genital lesion was present, an additional swab specimen was also collected from it for HSV testing. Specimens were stored at 4°C until they were transported to the laboratory within the next 24 h. Subsequent to specimen collection, participants received routine clinical care and counseling per STD clinic protocol

Herpes laboratory proceduresTesting for direct evidence and serological evidence of GH infections was conducted in the laboratory of R.W. and F.L

Culture for HSV was performed for each swab specimen, using BSC-1 and A549 cells in supplemented MEM. Cultures were observed daily for 1 week for cytopathic effects. Virus isolated in culture was typed by fluorescence microscopy using the Syva Microtrak HSV 1/HSV 2 typing test (Behring Diagnostics) [19]

PCR was performed on pooled aliquots created from viral transport media from all specimens collected from each participant. Specimen DNA was extracted using Qiagen extraction and concentration kits. Extracted specimens (10 μL) were then combined with a mastermix preparation containing 20 μL of distilled water, 20 μL of Eppendorf Hot MasterMix (2.5×), and 0.2 μL of primer for the polymerase (pol) marker, as has been described elsewhere [25]. These preparations were then amplified on a Perkin Elmer GeneAmp PCR System 2400 (95°C for 30 s; followed by 45 cycles composed of 94°C for 20 s, 64°C for 20 s, and 72° for 20 s; and then finishing with 74°C for 3 min). Positive, blank (both extraction and mastermix preparation), and delta controls were utilized to examine for evidence of contamination or inhibition. Electrophoresis for identification of target bands was then conducted on agarose gels. Specimens positive for HSV DNA were confirmed by PCR using a second primer for glycoprotein B (gpB) in a similar fashion; if a specimen was found to be sequentially positive by use of both the pol and gpB primers, the pol PCR product was then analyzed by electrophoresis after the application of Cfo restriction enzyme to determine viral type

HSV serological testing was performed as outlined in the manufacturer’s instructions, using the HerpeSelect 1 ELISA IgG and HerpeSelect 2 ELISA IgG assays (Focus Technologies) to determine the presence of antibodies to HSV-1 and HSV-2. As per those instructions, specimens with OD values ⩾1.1 were considered to be positive, and those with OD values <0.9 were considered to be negative; for those with values in the indeterminate range of 0.9–1.1, testing was repeated. If optical density values remained in the indeterminate range on repeat testing, the specimens were considered to be negative. If the optical density values in the repeat assay did not fall in the indeterminate range, the second value was used in assigning serological status

Clinical definitions of HSV infectionsPatients’ HSV infections were categorized in a way similar to that used in a prior study [3]. Primary genital infections were defined as those in which either HSV-1 or HSV-2 was recovered from a genital lesion and in which type-specific serological tests for both HSV-1 and HSV-2 were negative. Nonprimary first-episode genital infections were defined as those in which either HSV-1 or HSV-2 was recovered from a genital lesion and in which the serological test for the viral type isolated was negative and the serological test for the other viral type was positive. Recurrent disease was defined as that in individuals who had virus isolated from a lesion and whose corresponding serological test was positive. Asymptomatic viral shedding was defined as the detection of virus in the absence of lesions or symptoms. Latent genital infection was defined only for HSV-2; individuals with this infection type had antibodies to HSV-2 in the absence of viral detection from a genital site

Data management and statistical analysisQuestionnaire and laboratory data were entered into a Microsoft Access database with appropriate checks to ensure accuracy. Analyses were conducted with SAS software (version 9.0; SAS Institute). Univariate analyses were conducted using Student’s t test and Fisher’s exact test; logistic regression was used for the multivariate analysis

Results

Study populationOverall, 518 men were enrolled. Of the 516 men with data available, 90% (n=465) were African American (median age, 25 years), 76% had completed high school or had a GED, and 69% were currently employed. Reflecting the nature of our clinic population [20] as well our exclusion criterion regarding MSMs, the entire population was heterosexual; participants reported a median of 1 sex partner during the 30 days preceding enrollment, 2 during the past 6 months, and 15 over the course of their lifetimes. The median age of self-reported sexual debut was 15 years (table 1)

Sixty-six percent (n=342) of participants consented to HIV testing; the HIV seroprevalence in this population was 0.29% (n=1). When asked about their STD history, 5.6% of participants reported a history of GH, and an additional 17% reported a history of recurring genital lesions; both findings are suggestive of possible prior GH. In addition, 15.5% presented with complaints of genital ulceration at the time of the study visit, one-third of whom reported a history of similar episodes. Among participants who were HSV-2 seropositive, men with antibodies to HSV-1 were no less likely to report a history consistent with GH than were their HSV-1–negative counterparts

HSV type–specific serological testingOverall, 45% (n=233) of participants had serological evidence of HSV-2 infection, and 315 (61%) had antibodies to HSV-1; 28% were seropositive for both HSV-1 and HSV-2, and 17% were seropositive for HSV-2 alone. An additional 33% of participants were HSV-1 seropositive only, and 22% were seronegative for both viruses (table 2)

Factors associated with HSV-2 infection (n=244)In univariate analyses, African American men were 2.2 times more likely to be infected with HSV-2 than were white men (P=.01). Other identified associations with HSV-2 seropositivity included older age (P<.0001), self-reported history of prior intranasal cocaine use (P=.0006), history of encounters with commercial sex workers (P=.02), and a history of incarceration (P=.02). Men with evidence of HSV-2 infection were less likely to be single and never married than were men without HSV-2 infection (P=.004). The groups did not differ with respect to educational level; employment status; tobacco, alcohol, or injection drug use; reported condom use; health insurance status; or HSV-1 serostatus. In the multivariate analysis including variables with P<.15 in the univariate analyses described above, only older age, African American race, and the number of sex partners during the past 30 days remained significantly associated with evidence of HSV-2 infection (table 3)

With respect to number of sex partners, there were trends toward HSV-2–seropositive individuals reporting more lifetime sex partners (42 vs. 28; P=.06) and more sex partners during the past 30 days (1.67 vs. 1.44; P=.02) than their seronegative counterparts. When stratified by age and number of lifetime sex partners, there were significant trends for HSV-2 seropositivity and increasing age as well as increasing numbers of lifetime sex partners. Furthermore, we noted a previously unreported association between HSV-2 seropositivity and total years of sexual experience, defined as current age minus age of sexual debut (figure 1)

Spectrum of genital HSV infectionsIn total, 247 participants (48%) had evidence of GH by serological and/or viral testing—241 with HSV-2 and 9 with HSV-1 (table 4). Of the 52 men with HSV detected in genital swab specimens, 43 were positive for HSV-2 (82.7%) and 9 were positive for HSV-1 (17.3%). Again, on the basis of both serological and viral test results, among the 241 (47.1%) men with evidence of HSV-2 infection, 6 (2.4%) had symptomatic primary infection, 2 (0.8%) had symptomatic nonprimary, first-episode genital infection, 27 (11.1%) had recurrent disease, 8 (3.3%) were asymptomatically shedding virus, and 198 (82.2%) had only antibody evidence of infection. Twenty-one (10.6%) of these 198 men reported a history of GH, and 60 (30%) reported a history of GH or recurring genital lesions

Thirty-three of the 35 men with symptomatic HSV-2 infection at the time of the study visit had descriptions of their genital lesions available. Most (25/33) had multiple lesions (range, 2–10), from pustules to vesicles to ulcers to crust. All lesions in patients testing positive for HSV-2 were located on the penis, with the exception of 2 participants, one of whom had a scrotal ulcer and the other of whom had a lesion at an unspecified location

Of the 9 men with HSV-1 detected from their genitalia, 2 had primary genital infection, 1 had nonprimary, first-episode genital infection, 1 had a recurrent disease, and 5 were asymptomatically shedding virus. Three were also seropositive for HSV-2. None of these 9 individuals reported a history of GH, and only 1 reported a history of recurring genital lesions. Of the 4 symptomatic men shedding HSV-1, all presented with genital ulcers located on the penis (1 on the shaft, 2 on the coronal ridge, and 1 on the glans); 1 participant also had vesicles noted. Three (75%) of the 4 had multiple lesions. There were no significant differences with respect to age or race between men who had HSV-1 isolated and those who had HSV-2 isolated

Culture versus PCR for HSV detectionAs a component of this study, we also compared PCR and culture as viral detection techniques. The overall viral shedding rate among men (n=247) with evidence of GH (HSV-1 or HSV-2) was 21.1%, whereas the asymptomatic shedding rate among this same group was 5.2%. When measured by culture, 88% (22/25) of positive HSV-2 cultures among those with evidence of HSV-2 infection occurred in symptomatic patients, for an overall shedding rate of 10.4% and an asymptomatic shedding rate of 1.25%. All isolates that were culture positive were PCR positive, but in approximately half of the PCR-positive patients, cultures were negative. Specifically, when PCR was utilized, 81% (35/43) of the HSV2 viral shedding detected was among symptomatic individuals, yielding an overall shedding rate of 17.8% and an asymptomatic shedding rate of 3.3%. The sensitivities of culture for HSV-1 and HSV-2 were .22 and .58, respectively, when compared with PCR. For primary and nonprimary, first-episode genital HSV-2 infections, the sensitivity of culture was significantly better, at 88%

Coexisting STDsTwenty-five percent of the participants were given a clinical diagnosis of nongonococcal urethritis at the study visit. By use of nucleic acid amplification testing, 19% of participants tested positive for gonorrhea, and 19% tested positive for chlamydia. Individuals with evidence of HSV-2 infection were no more likely to have these other STDs than were those without HSV-2 infection. One incident case of HIV infection was identified in a patient presenting with a severe recurrence of HSV-2 infection

Discussion

Our data confirm that African American men are approximately twice as likely as white men to be infected with HSV-2, indicating this group as an important subpopulation for study as part of efforts to understand the clinical epidemiology of GH. Over the past 15 years, 3 multicenter HSV-2 seroprevalence studies targeting different populations have documented this pattern of ∼2-fold increased risk of HSV-2 infection among African American men [1, 2, 4]. First, NHANES III (Third National Health and Nutrition Examination Survey), a study of a representative sample of the noninstitutionalized civilian US population, sampled 6407 men between 1988 and 1994, 1798 of whom were African American, and documented overall seroprevalences of 17.8% among all men, 14.9% among white men, and 34.7% among African American men [1]. Second, between 1993 and 1996, Project RESPECT, a multicenter study of STD clinic attendees in 5 US cities, found that 32% (n=2348) of male participants had HSV-2 antibodies—again, with African American men (39%) nearly twice as commonly infected as their white counterparts (20%) [4]. More recently, in 2004, a study by Leone et al. [2] included 2719 men attending suburban primary care sites in 6 major US cities; there, they found that, overall, 22% of men had serological evidence of HSV-2 infection, with African American men (43%; n=322) disproportionately infected in comparison with white men (18%; n=2099) (V. Williams, personal communication). Thus, although our study was conducted in an urban STD clinic, these other studies, conducted in different settings and using different methodologies, have consistently reported a relatively high burden of HSV-2 infection among African American men, irrespective of the venue of care. In addition, the similarity of HSV-2 seroprevalence (46%) in the present study and the populations described above suggests a degree of generalizability for our data

Our study adds to the limited published data regarding patterns of genital HSV shedding among men. Except for 2 reports from Sweden describing asymptomatic shedding among <50 predominantly white MSMs [21, 22], most published data regarding genital HSV shedding in men have been reported over the past decade by an accomplished group of investigators at the University of Washington [7–12]. Focusing on careful longitudinal characterization of GH—again, among relatively small cohorts of predominantly white HIV-positive or HIV-negative MSMs—they have noted the following findings. First, although the penis is the most common site of shedding among heterosexual men, the most common site of HSV shedding for MSMs, regardless of HIV status, is from the perianal area. Second, HIV-positive MSMs (9.7%) more often shed from the genital area than do HIV-negative MSMs (3.1%). Third, MSMs are more likely than heterosexual men to have HSV-1 isolated from genital sites when lesions are present, although HSV-2 remained the predominant type isolated overall. Fourth, overall shedding rates varied from 3% to 9%, and asymptomatic shedding rates varied from 1% to 3%; the exact rates were largely dependent on the detection technique used and the immune status of the participants. Our data extend these findings, showing rates of overall and asymptomatic HSV-2 shedding from penile sites in predominantly immunocompetent heterosexual African American men that are similar to those seen when sampling both perianal and penile sites in MSMs

Our data also provide a conservative estimate of the number of genital HSV-1 infections and suggest that genital HSV-1 infection is common. Compared with Lafferty et al.’s study, in which 14.6% of primary and nonprimary, first-episode GH and 7.4% of recurrent GH seen among 600 heterosexual males was caused by HSV-1 [12], our study found that a substantially higher percentage (3/11 [27.3%]) of initial episodes were caused by HSV-1, although only 1 (3.6%) of 28 recognized episodes of recurrent disease were caused by HSV-1. At the same time, of 13 asymptomatic men with swab specimens positive for HSV, 5 (38%) were HSV-1 positive. Because prior studies suggest that asymptomatic viral shedding is less common among persons with GH caused by HSV-1, our data imply that genital HSV-1 infections may be more common than has been previously appreciated. This inference is further supported by other recent studies from Great Britain, Israel, and a US university student clinic, suggesting that an increasing proportion of genital HSV isolates are HSV-1 [15–17, 23, 24]. Distinct from these studies however, our data did not suggest that younger age predicted a higher proportion of HSV-1 isolation

Finally, our study confirms that PCR is more sensitive than culture for detection of genital HSV infection, especially in the absence of clinical signs. Recently, Wald et al. summarized their experience comparing culture and PCR for 25,000 mucosal samples collected from a predominantly (86%) white sample of 159 men, 77 (48%) of whom were HIV antibody positive [18]. They reported detecting HSV in 3.9% of samples by culture and in 10% of samples by PCR, with a ratio of PCR positivity to viral culture positivity of 4.0:1. This ratio was not significantly different than that observed in women or when men were stratified by HIV status; however, the ratio was higher when lesions were absent (5.1:1) than when lesions were present (3.1:1). In our study, the ratios were similar although less dramatic, with an overall ratio of 2:1 (1.25:1 when symptomatic lesions were present and 2.4:1 for asymptomatic shedding). Our findings of relatively increased sensitivity of culture in persons with new infections, as well as in persons infected with HSV-2 in comparison with those infected with HSV-1, suggest that, on the basis of prior reports of higher viral loads being associated with these categories of infection, local viral load likely plays an important role in explaining the difference in sensitivity. Our findings reaffirm that PCR is the method of choice for genital HSV detection

Several limitations of this study should be acknowledged and considered. First, these data were collected from a single STD clinic in the southeastern United States and, thus, may not be generalizable to other populations of African American men. Nonetheless, as emphasized above, the similar rates of HSV-2 infection seen in African American men in our study and in men seen at suburban primary care clinics across the United States suggest that this may not be the case. Another limitation of our study is that we did not include the perianal area as one of our sampling sites, thus potentially underestimating the amount of shedding in these men. Prior studies, however, have indicated that perianal shedding is relatively uncommon among heterosexual men. Nevertheless, we recommend that future studies formally examine the role of perianal shedding in heterosexual men. Additionally, future studies should consider viral DNA quantification as an additional measurable factor to account for potential difference in rates of infection seen between populations. Although our study certainly defines the spectrum of GH shedding among these men, it is cross-sectional in nature; future studies focusing on longitudinal cohort data from populations of African American men with known genital HSV infections would complement prior studies conducted by our colleagues at the University of Washington

In describing the spectrum of GH among the largest sample of men to date by use of sensitive techniques for viral detection, we found that HSV infections, particularly HSV-2 infections, are common in this previously understudied population of heterosexual men, although genital HSV-1 infections represented a nontrivial proportion of the genital HSV infections detected. Most infections were not recognized by participants. Increasing age, African American race, increasing number of sex partners, and increasing years of sexual experience were positively associated with rates of HSV infection