Nationwide serosurveys indicate a declining herpes simplex virus type 2 (HSV-2) and type 1 (HSV-1) prevalence in the United States. Among persons aged 14 to 49 years, HSV-2 seroprevalence has decreased from 18.0% in 1999 to 2000 to 12.1% in 2015 to 2016.1 The prevalence of HSV-1 has also been declining, particularly in adolescents aged 14 to 19 years, who had a 23% decline from 39.0% to 30.1% between 1999 and 2010.2
In parallel with changing HSV seroepidemiology, the etiology of first-episode anogenital herpes also seems to be shifting. High-income countries have reported an increase in the proportion of first-episode anogenital herpes due to HSV-1, from 29% in the 1990s to 42% in the 2000s.3–5 In particular, HSV-1 is the most common cause of first-episode genital herpes infections in studies conducted among young adults aged 18 to 24 years, accounting for up to 80% of first-episode genital herpes infections among university students.6,7 Genital HSV-2 infection is associated with a more severe infection as compared with genital HSV-1, with substantially more frequent recurrences. Therefore, although HSV-1 is emerging as a notable cause of first-episode genital herpes, HSV-2 continues to cause a higher proportion of HSV-associated genital ulcer disease than does HSV-1.8
In an earlier study, approximately 20% of patients with first-episode genital herpes evaluated at the Public Health – Seattle & King County (PHSKC) Sexually Transmitted Disease (STD) Clinic between 1993 and 1997 had laboratory-documented genital HSV-1 infection.9 To evaluate time trends in first-episode genital herpes infection, we assessed the number and proportion of diagnoses of first-episode genital herpes due to HSV-1 and HSV-2 between 1993 and 2014 in the PHSKC STD Clinic. Among those with first-episode genital herpes, we also assessed risk factors associated with first-episode genital HSV-1 versus HSV-2 infection.
MATERIALS AND METHODS
The PHSKC STD Clinic provides confidential STD and HIV evaluation, screening, testing, and treatment on a sliding fee basis. Persons presenting with a genital ulcer and suspected genital herpes typically have their lesions swabbed for HSV. Patients who visited between January 1, 1993, and December 31, 2014, and who met any of the following criteria were included: (1) diagnosis of genital ulcers, (2) collection of a swab for viral culture, and (3) physical examination findings consistent with genital herpes. Those with virologic evidence of genital herpes (positive genital HSV culture) and no documented or self-reported history of lesions were included for further analysis.
Demographic and Sexual History Collection
The PHSKC STD Clinic uses standard collection instruments for all clinical encounters, including demographic information (age, self-identified sex, race/ethnicity), detailed sexual history (sex of sex partners, number of lifetime sex partners), sexual practices (insertive and receptive oral, vaginal and anal sex), physical examination findings, and laboratory testing. Until October 2010, clinicians assessed and recorded all clinical and behavioral information using face-to-face interviews entered into an electronic medical record database.10 After October 2010, an expanded set of behavioral and self-reported clinical information was collected directly from patients using a computer-assisted self-interview.11–13 If patients did not complete the computer-assisted self-interview because of language barriers or logistical issues, behavioral measures were collected by clinicians.
Samples for viral culture underwent shell vial centrifugation in MRC-5 cells and were processed using Trinity MicroTrak monoclonal antibody reagents as per the manufacturer's instructions.14 Cultures were typed as HSV-1 or HSV-2 using type-specific monoclonal antibodies.
We used linear regression analysis to examine time trends in number of first-episode genital HSV infections, including overall numbers of infections of each viral type per year, and stratified by sex, race, and sex of sex partners. To determine risk factors associated with genital HSV-1 infection among those with first-episode genital herpes, we included sex, age, race/ethnicity, and sex of sex partners a priori. We performed univariate and multivariate analyses using Poisson regression models. We included variables from the univariate model in our multivariate model, removing those with P > 0.05 with backward elimination. Two-sided P values less than 0.05 were considered significant.
To address the potential influence of the large number of uncultured genital lesions on the observed time trends in HSV characterization, we applied two sensitivity analyses post hoc. In each, we assumed that a proportion of the uncultured samples from persons who otherwise met the criteria for first-episode genital herpes (no self-reported or documented history) were positive for HSV-1 or HSV-2, calculated the number of hypothetical cases of each infection based on these proportions, and then calculated total numbers of first-episode infections by adding the hypothetical cases to the observed first episode cases of each viral type. In the first sensitivity analysis, we assumed that all uncultured samples for patients with first-episode genital lesions were positive for HSV-1 or HSV-2 at the same frequency as samples collected that same year. These hypothetical new cases were added to the observed cases for each year and plotted over time. In a second sensitivity analysis, we remained agnostic to any potential time trend and applied to the uncultured samples the HSV-1 and HSV-2 culture positivity frequencies obtained over all samples over the 22-year study period. The total cases for each virus per year were then plotted over time.
This study was approved by the University of Washington Human Subjects Division.
Between 1993 and 2014, 52,030 patients had a diagnosis or physical findings of genital ulcers (Fig. 1). We excluded 8195 (15.8%) persons who had recurrent genital HSV. Of the remaining 43,835 patients presenting with no documentation of a history of genital ulcers, we excluded an additional 30,651 (70.0%) who did not have cultures collected and 9929 (22.7%) persons who had cultures that were negative for HSV. Of the remaining 3255 (7.1%) patients with positive HSV cultures, we excluded 83 with positive HSV cultures from the mouth, 2 with both HSV-1 and HSV-2, and 105 that were not typed. This left 3065 patients (94% of the culture-positive samples) with first-episode genital herpes documented by a virologic test for the final analysis. Of these, 1022 (33.3%) first-episode genital herpes infections were due to HSV-1 and 2043 (66.7%) were due to HSV-2.
Among the 3065 persons with culture-proven first-episode genital herpes infection, the median age was 28 years (interquartile range, 24–36 years) and 1154 (37.7%) were women. Patients identified as non-Hispanic white (1875 [61.2%]), non-Hispanic black (679 [22.2%]), Asian (142 [4.6%]), Hispanic (73 [2.4%]), Pacific Islander (26 [0.8%]), Native American (44 [1.4%]), or multiracial (137 [4.5%]) (Table 1). Of 1,154 women, 919 (79.6%) reported only male sex partners, 12 (1%) reported female partners only, and 51 (4.4%) reported female and male partners; 172 (14.9%) did not indicate sex of sex partners. Of the 1911 men, 1241 (64.9%) reported female sex partners only, 295 (15.4%) reported male partners only, 58 (3%) reported male and female sex partners, and 317 (16.6%) did not indicate sex of sex partner.
Temporal Trends in First-Episode Genital Herpes
A median of 93 persons were diagnosed as having first-episode genital HSV-2 per year (range, 35–208). The number of patients with genital HSV-2 per year decreased steeply from 208 in 1993 to 72 in 2000, and then further decreased to 35 in 2014 (Fig. 2A), with an average reduction of 5.5 HSV-2 infections per year (95% confidence interval [CI], −6.9 to −4.1; P < 0.0001). Over the same period, the number of HSV-1 infections was relatively constant, with a median of 46 (range, 28–68), an average change of 0.2 per year (95% CI, −0.4 to 0.9; P = 0.52). Each decade, the proportion of first-episode genital herpes due to HSV-1 increased by 52% (relative risk [RR], 1.52; 95% CI, 1.38–1.66; P < 0.0001). Inversely, the proportion of first-episode genital herpes due to HSV-2 decreased significantly by 20% per decade (RR, 0.80; 95% CI, 0.77–0.84; P < 0.0001).
When stratified by sex, 673 (65.9%) of HSV-1 and 1238 (60.6%) of HSV-2 infections were in men (Table 1). A marked decline in HSV-2 diagnoses in both men and women was found over time (Fig. 2B). We also stratified the number of first-episode genital herpes diagnoses by self-reported race and ethnicity. We found that 772 (75.5%) persons with HSV-1 infections and 1103 (54.0%) persons with HSV-2 infections were non-Hispanic white and 82 (8%) persons with HSV-1 infections and 597 (29.2%) persons with HSV-2 infections were non-Hispanic black (Table 1). There was a significant decrease over time in the number HSV-2 infections in both groups, similar to the overall trend noted, with stable, low numbers of HSV-2 infections among other groups (Supplemental Fig. 1A, http://links.lww.com/OLQ/A422). In comparison, there was no significant change in the number of persons who acquired HSV-1 when stratified by race/ethnicity—following the overall HSV-1 trend over time—with a median of 35 (range, of 21–52) white persons diagnosed as having genital HSV-1 per year (Supplemental Fig. 1B, http://links.lww.com/OLQ/A423). When stratifying by sex of sex partners among men, first-episode genital HSV-2 diagnoses fell sharply over time among MSW and remained relatively constant among men who have sex with men (MSM; Fig. 2C). There were not sufficient numbers of women who have sex with only women to perform similar analyses in women.
Sexual Practices Among Those With First-Episode Genital Herpes
Of 650 women with data regarding oral sex, 627 (96.5%) had received oral sex; and of 967 with data regarding vaginal sex, 965 (99.8%) had vaginal sex. Similarly, of 665 men who reported their receptive oral sex practices, 660 (99.2%) had received oral sex; and of 614 men who reported their insertive vaginal sex practices, all (100%) had insertive vaginal sex. Of the 353 men who reported male sex partners and acquired genital HSV-1, 74 (20.9%) had insertive anal sex and 63 (17.8%) had receptive anal sex.
Risk Factors Associated With Genital HSV-1 Among Those With First-Episode Genital Herpes
We performed univariate regression to determine correlates of acquisition of genital HSV-1 versus HSV-2 among those with first-episode genital herpes. Younger age (<30 years) was associated with 1.19 times the risk of having genital HSV-1 (95% CI, 1.07–1.32; P = 0.0011; Table 2). Men were significantly more likely to acquire HSV-1 than women (RR, 1.16; 95% CI, 1.05–1.29; P = 0.005). Compared with black persons, white persons (RR, 3.41; 95% CI, 2.83–4.11; P < 0.0001), Asian persons (RR, 3.91; 95% CI, 3.00–5.98; P < 0.0001), and Pacific Islanders (RR, 4.45; 95% CI, 2.81–7.09, P < 0.001) were more likely to acquire HSV-1. Men who have sex with men had an increased risk of acquiring HSV-1 compared with men who reported only heterosexual contact and women (RR, 1.79; 95% CI, 1.57–2.03; P < 0.0001). There was a 53% increased risk of being diagnosed as having genital HSV-1 as compared with HSV-2 each decade after 1993 (RR, 1.53; 95% CI, 1.41–1.65; P < 0.0001).
In a multivariate model, women no longer had a significantly lower risk of genital HSV-1 than did men, but persons who were <30 years of age (RR, 1.37; 95% CI, 1.21–1.55; P < 0.0001), white (RR, 3.16; 95% CI, 2.57–3.88; P < 0.0001), Asian (3.5; 95% CI, 2.52–4.68; P < 0.0001), and Pacific Islander (RR, 4.08; 95% CI, 2.52–6.62; P < 0.0001) and MSM (RR, 1.48; 95% CI, 1.29–1.69; P < 0.0001) remained significantly more likely to have HSV-1 than HSV-2 (Table 2). The risk of an HSV-1 diagnosis increased by 52% per each additional decade (RR, 1.52; 95% CI, 1.38–1.66; P < 0.0001) compared with HSV-2. Information on sexual behavior was missing for a large proportion of persons, so these measures were not included in multivariate regression models.
We found that the number of patients presenting to the clinic with genital ulcers also declined over time. In addition, most patients (70%) presenting with symptoms of genital herpes did not have HSV cultures performed. Over time, we found that there was less reliance on virologic diagnosis for people presenting with genital ulcers. For example, in 1993, 38% of people presenting with genital ulcers (1154/3039) had cultures obtained, whereas in 2014, 18% (205/1125) had their samples cultured. The proportion of people whose samples were cultured decreased by 2.9% every year (95% CI, 2.7%–3.1%, P < 0.001). To establish whether there was bias in who had cultures collected, we determined whether demographic factors were associated with having cultures collected. We found that women were more likely than men to have their genital lesions cultured (48% versus 23%), but that culture rates were not significantly different by age, race, or sex practices. After adjustment for sex, the proportion of patients who had their samples cultured decreased by 2.1% every year (95% CI, 1.9%–2.3%, P < 0.001). Therefore, we conducted sensitivity analyses to attempt to understand whether our findings represented epidemiologic as opposed to ecologic trends. In the first sensitivity analysis—in which we assumed that all uncultured samples in first-episode genital lesion patients were positive for HSV-1 or HSV-2 at the same frequency as samples sent for analysis in that same year—on average, approximately 75% of samples tested were negative for HSV, and the annual proportion of positive samples ranged from 3% to 18% for HSV-1 and from 14% to 18% for HSV-2. In this analysis, the number of HSV-2 diagnoses decreased from 552 in 1993 to 181 in 2014, whereas the number of HSV-1 diagnoses seemed to increase moderately from 114 in 1993 to 170 in 2014 (Fig. 3A). In the second sensitivity analysis, we applied the same hypothetical culture positivity rate to untested samples from all years (8% contained HSV-1 and 16% HSV-2). In this model, the HSV-2 diagnoses similarly decreased from 504 in 1993 to 180 in 2014, and HSV-1 diagnoses decreased from 191 in 1993 to 104 in 2014 (Fig. 3B).
In this analysis of time trends in first-episode genital herpes infection diagnoses in an urban STD clinic, we found that the number of persons diagnosed as having culture-proven first-episode genital HSV-2 infection declined significantly from 1993 to 2014, whereas first-episode genital HSV-1 diagnoses have remained fairly stable. We identified several demographic factors associated with first-episode genital HSV-1, including younger age, persons who were white or Asian or Pacific Islander, and MSM. We found that the increasing proportion of first-episode genital HSV-1 infection was due to a decrease in diagnoses of genital HSV-2 rather than an increase in HSV-1 diagnoses. These findings are consistent with those from the National Health and Nutrition Examination Surveys, which has shown a decline in HSV-2 seroprevalence from a high of 21% to 16.2% in 2010, with a further decline to 12.1% in 2015 to 2016.1 In addition, a study of pregnant women in Seattle showed a decrease in HSV-2 seroprevalence from 30.1% to 16.3% between 1989 and 2010.15,16
Acquisition of HSV-1 during childhood has decreased in the United States, resulting in a 29% decline in HSV-1 seroprevalence among adolescents between 1999 and 2010.17,18 Although this places adolescents and young adults initiating sexual activity at risk of acquiring genital HSV-1, the number of genital HSV-1 diagnoses did not increase over time, suggesting that whatever factors have led to the decrease in the number of HSV-2 diagnoses may also be preventing an increase in HSV-1 diagnoses despite an increased susceptible population.
The drivers of these changing trends in the epidemiology of first-episode genital herpes in King County are unknown. We hypothesize that there may be older age at sexual debut, shifts in sexual practices, or improved awareness and increased use of interventions known to decrease the risk of HSV-2 transmission. However, these findings are in stark contrast to trends in reportable sexually transmitted infections (STIs), such as syphilis, gonorrhea, and chlamydia, which have surged nationally and in King County over this period.19,20 Although daily oral HIV preexposure prophylaxis with tenofovir disoproxil/emtricitabine is associated with a 50% reduction in HSV-2 acquisition in some studies, this intervention was not Food and Drug Administration approved until 2012 and consequently would not affect these trends.21 It therefore seems likely that the risk of transmission has decreased, possibly through the use of antiviral medication, particularly suppressive therapy, which decreases the risk of subclinical viral shedding and transmission.22 Although antiviral medications have become increasingly available over the time frame that the decline in HSV-2 has been noted, there are no data to confirm this hypothesis. We do not have a national surveillance system in place for first-episode genital HSV infection. To better understand the trends in genital HSV infections, it would be of interest to study the epidemiology among high-risk populations, such as those presenting with other STIs or taking preexposure prophylaxis to prevent sexual HIV acquisition. In addition, if available diagnostics were better used in a systematic way, we could gain additional insights into the changing patterns of this chronic and common STI.
Our results are similar to other recent studies in demonstrating risk factors for HSV-1 compared with HSV-2 infection. We found an increased risk of genital HSV-1 infection among persons <30 years old, white, and MSM, similar to other studies that have found genital HSV-1 infection associated with white MSM.23,24 Other studies have found a higher incidence of HSV-1 in women than in men.23 These associations suggest that specific populations are at risk of and may benefit from targeted prevention messages for genital HSV-1 infections, including increasing awareness that genital HSV-1 infection can be transmitted through oral, vaginal, and anal sex. The almost universal practice of receptive oral sex in this study hampered analyses of associations of this behavior with acquisition of HSV-1 and HSV-2.
This study had several strengths, including a large data set with many patients, accessed through the thorough electronic record maintained by the PHSKC STD Clinic. However, a significant number of responses regarding sexual practices of patients were missing, leading to inability to evaluate sexual behaviors as risk factors in the multivariate analysis. In addition, many people presented with genital ulcers but did not have cultures obtained, and the number of cultures obtained over time declined, which may have biased the results. However, the sensitivity analyses demonstrated that even when assuming a proportion of these samples were positive for HSV-1 or HSV-2, a strong downward trend was still seen in the number of HSV-2 diagnoses. This persistent decrease is partly due to the fact that the overall number of persons seen with genital ulcers is also decreasing over time, supporting the notion that genital herpes is decreasing in this population. Finally, there may be other factors leading to the trends that we identified. For instance, an increasing number of patients may seek clinical care in locations other than the PHSKC STD Clinic, resulting in decreasing trends in the number of patients diagnosed as having genital herpes at this location.17 It is difficult to know whether these results would be generalizable to other STI clinics, particularly those clinics that serve more women. However, a recent study published out of Australia showed similar findings,25 with an increasing proportion of first-episode genital herpes diagnoses due to HSV-1, suggesting that this is not simply a local trend.
Interestingly, we found that fewer cultures were collected from people presenting with genital ulcers over time. In addition, we found that most culture samples (9927/13,184 [75%]) collected were negative for HSV. We now have greater availability of polymerase chain reaction/nucleic acid amplification test for detection of HSV from lesions, which is approximately 4-fold more sensitive than culture.26 If polymerase chain reaction/nucleic acid amplification test had been available, we would have undoubtedly detected more cases of genital HSV. However, for a first-episode genital lesion, we would expect that culture would be equally likely to detect HSV-1 and HSV-2, and the results of the sensitivity analysis support that HSV-2 diagnoses are declining. It would be of interest to conduct similar studies in the future using the more sensitive, modern diagnostic tools that are now available.
In summary, we observed a significant decrease in first-episode genital HSV-2 diagnoses over the last 2 decades, with a stable number of first-episode genital HSV-1 diagnoses, resulting in an increased proportion of first-episode genital herpes attributed to HSV-1. Understanding changing epidemiology in genital herpes infection may inform prevention strategies and counseling messages, including increasing awareness that genital HSV-1 infection can be transmitted through oral, vaginal, and anal sex; that people who have not previously acquired oral HSV-1 can be at risk of acquiring HSV-1 in the genital tract; and the importance of typing first Episode Genital HSV infections.
1. McQuillan G, Kruszon-Moran D, Flagg E, et al. Prevalence of herpes simplex virus type 1 and type 2 in persons aged 14–49: United States, 2015–2016. NCHS Data Brief 2018; 304.
2. Bradley H, Markowitz L, Gibson T, et al. Seroprevalence of herpes simplex virus types 1 and 2—United States, 1999–2010. J Infect Dis 2014; 209:325–333.
3. Ryder N, Jin F, Mcnulty A, et al. Increasing role of herpes simplex virus type-1 in first episode anogenital herpes in heterosexual women and younger men who have sex with men, 1992–2006. Sex Transm Infect 2009; 85:416–419.
4. Wald A. Genital HSV-1 infections. Sex Transm Infect 2006; 82:189–190.
5. Scoular A, Norrie J, Gillespie G, et al. Longitudinal study of genital infection by herpes simplex virus type 1 in western Scotland over 15 years. BMJ 2002; 324:1366–1367.
6. Horowitz R, Aierstuck S, Williams E, et al. Herpes simplex virus infection in a University health population: clinical manifestations, epidemiology, and implications. J Am Coll Health 2010; 59:69–74.
7. Roberts C, Pfister J, Spear S. Increasing propotion of herpes simplex virus type 1 as a cause of genital herpes infection in college students. Sex Transm Dis 2003; 30:797–800.
8. Lowhagen G, Tunback P, Andersson K. First episodes of genital herpes in a Swedish STD population: A study of epidemiology and transmission by the use of herpes simplex virus (HSV) typing and specific serology. Sex Transm Infect 2000; 76:179–182.
9. Lafferty W, 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; 181:1454–1457.
10. Khosropour C, Dombrowski J, Swanson F, et al. Trends in serosorting and the association with HIV/STI risk over time among men who have sex with men (MSM). J Acquir Immune Defic Syndr 2016; 72:189–197.
11. Chambers L, Manhart L, Katz D, et al. Evaluation of an automated express care triage model to identify clinically relevant cases in a sexually transmitted disease clinic. Sex Transm Dis 2017; 44:571–576.
12. Chambers L, Manhart L, Katz D, et al. Comparison of algorithms to triage patients to express care in a sexually transmitted disease clinic. Sex Transm Dis 2018; 45:696–702.
13. Khosropour C, Dombrowski J, Hughes J, et al. Evaluation of a computer-based recruitment system for enrolling men who have sex with men into an observational HIV behavioral risk study. Am J Epidemiol 2016; 184:477–483.
14. Trinity Biotech. MicroTrak HSV-1/HSV-2 Culture Identification/Type Testing. 2014.
15. Delaney S, Gardella C, Saracino M, et al. Seroprevalence of herpes simplex virus type 1 and 2 among pregnant women, 1989–2010. JAMA 2014; 312:746–747.
16. Patton M, Bernstein K, Liu G, et al. Seroprevalence of herpes simplex virus types 1 and 2 among pregnant women and sexually active, nonpregnant women in the United States. Clin Infect Dis 2018; 67:1535–1542.
17. Auslander B, Biro F, Rosenthal S. Genital herpes in adolescents. Semin Pediatr Infect Dis 2005; 16:24–30.
18. Ayoub H, Chemaitelly H, Abu-Raddad L. Characterizing the transitioning epidemiology of herpes simplex virus type 1 in the USA: Model-based predictions. BMC Med 2019; 17:57.
19. CDC. 2018; Available at: https://www.cdc.gov/std/stats17/toc.htm
. Accessed June 10, 2019.
20. Seattle & King County Public Health. 2015 King County Sexually Transmitted Diseases Epidemiology Report. 2015.
21. Celum C, Morrow R, Donnell D, et al. Daily oral tenofovir and emtricitabine-tenofovir preexposure prophylaxis reduces herpes simplex virus type 2 acquisition among heterosexual HIV-1–uninfected men and women: A subgroup analysis of a randomized trial. Ann Intern Med 2014; 161:1–15.
22. Corey L, Wald A, Patel R, et al. Once-daily Valacyclovir to reduce the risk of transmission of genital herpes. NEJM 2004; 350:11–20.
23. Cherpes T, Meyn L, Hillier S. Cunnilingus and vaginal intercourse are risk factors for herpes simplex virus type 1 acquisition in women. Sex Transm Dis 2005; 2:84–89.
24. Janier M, Scieux C, Meouchi R, et al. Virological, seriological and epidemiological study of 255 consecutive cases of genital herpes in a sexually transmitted disease clinic of Paris (France): A prospective study. Int J STD AIDS 2006; 17:44–49.
25. Durukan D, Fairley C, Bradshaw C, et al. Increasing proportion of herpes simplex virus type 1 among women and men diagnosed with first-episode anogenital herpes: A retrospective observational study over 14 years in Melbourne, Australia. Sex Transm Infect 2019; 95:307–313.
26. Wald A, Huang M, Carrell D, et al. Polymerase chain reaction for detection of herpes simplex virus (HSV) DNA on mucosal surfaces: Comparison with HSV isolation in cell culture. J Infect Dis 2003; 188:1345–1351.