Article Text

Original article
Risk factors for incident HSV-2 infections among a prospective cohort of HIV-1-discordant couples in China
  1. Yingying Ding1,
  2. Zunyou Wu2,
  3. Song Duan3,
  4. Keming Rou2,
  5. Yuecheng Yang3,
  6. Jibao Wang3,
  7. Meiyang Gao1,
  8. Runhua Ye3,
  9. Roger Detels4,
  10. Na He1
  1. 1Department of Epidemiology, School of Public Health, and the Key Laboratory of Public Health Safety of Ministry of Education, Fudan University, Shanghai, China
  2. 2National Center for AIDS/STD Control and Prevention (NCAIDS), Chinese Center for Disease Control and Prevention (China CDC), Beijing, China
  3. 3Dehong Prefecture Center for Disease Control and Prevention, Mangshi, Yunnan Province, China
  4. 4Department of Epidemiology, School of Public Health, University of California, Los Angeles, Los Angeles, USA
  1. Correspondence to Dr Na He, Department of Epidemiology, School of Public Health, Fudan University, Shanghai, 200032, China; nhe{at}shmu.edu.cn

Abstract

Objectives Identification of risk factors is essential for developing herpes simplex virus type 2 (HSV-2) prevention interventions that could also reduce HIV-1 transmission, particularly among HIV-1-discordant couples.

Methods A prospective cohort study was conducted among HIV-1-discordant couples from June 2009 to March 2011 in Yunnan province, China. 413 HIV-1-infected partners and 517 HIV-1-uninfected partners who were HSV-2 seronegative or equivocal at enrolment and who had a study partner completing the baseline survey and HSV-2 testing were included in the analysis.

Results HSV-2 incidence was 2.9 per 100 person-years (PY) for HIV-1-infected partners and 4.5 per 100 PY for HIV-1-uninfected partners. At least 36% of incident HSV-2 infections were from outside sexual partner. Among HIV-1-infected partners, multivariate analysis indicated that HSV-2 incidence was significantly higher among those with baseline equivocal HSV-2 result, having an initially HSV-2 seropositive or equivocal partner, reporting no sex with study partner and initiating antiretroviral therapy (ART) during follow-up. Among HIV-1-uninfected partners, multivariate analysis indicated that HSV-2 incidence was significantly higher among those having an initially HSV-2 seropositive partner and reporting sex with study partner ≥5 times/month, but was lower among those having a partner with baseline CD4+ count ≥350 cells/μL.

Conclusions Our findings underscore the importance of developing prevention and intervention programmes to reduce HSV-2 transmission among this population. The relationship between ART initiation and HSV-2 seroconversion requires further investigation.

  • HIV
  • CD4
  • EPIDEMIOLOGY (GENERAL)
  • HSV

Statistics from Altmetric.com

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Introduction

Herpes simplex virus type 2 (HSV-2) is one of the most common sexually transmitted infections (STIs) worldwide and the leading cause of genital ulcer disease (GUD).1 High HSV-2 rates have been observed among HIV-infected persons.2 ,3 There is a strong synergy between HSV-2 and HIV-1.4 ,5 Longitudinal studies have shown that prior HSV-2 infection increased the risk of acquiring and transmitting HIV-1,6 ,7 and HIV-1 infection was associated with a higher incidence of HSV-2.8 It has been suggested that controlling and preventing HSV-2 infection could potentially decrease HIV-1 transmission, particularly among HIV-1-discordant couples in which one partner is infected and the other is uninfected.

A recent longitudinal study has shown that incident HSV-2 infections were common among HIV-1-discordant couples.9 The Muiru et al9 study identified risk factors included being woman, having a partner who was initially HSV-2 seropositive or equivocal and reporting any unprotected sex with study partner during follow-up. Several studies have indicated that genital HSV-2 shedding among HIV-1-infected individuals was associated with a lower CD4+ count10–12 and increased after antiretroviral therapy (ART) initiation,13 ,14 indicating that immunological status and ART initiation of HIV-1-infected partners may have an impact on HSV-2 transmission among HIV-1-discordant couples. However, few observational studies have addressed this issue.

Therefore, in this study, we investigated HSV-2 incidence and associated risk factors among HIV-1-infected and HIV-1-uninfected partners in stable HIV-1-discordant heterosexual couples. The following risk factors were examined: sociodemographic characteristics, sexual behaviours, CD4+ counts, plasma HIV viral load and time of ART initiation for HIV-1-infected partners.

Methods

Study participants

This study used data from a prospective, longitudinal cohort study of HIV-1-discordant couples conducted from June 2009 to March 2011, in Dehong prefecture, Yunnan, as described elsewhere.15 HSV-2 prevalence at baseline has been reported previously.16 All known HIV-1-discordant couples in Dehong were invited to participate in the study between June 2009 and June 2010. All participants were asked to return for one follow-up visit 12 months after the baseline survey but before March 2011. Eligible participants were ≥18 years old and resided in the selected town. Written informed consent was obtained from all participants.

At baseline and at the follow-up visit, each participant was interviewed face to face by a trained health professional in a private place. Both HIV-1-infected and HIV-1-uninfected partners were asked about their demographic characteristics and sexual behaviours. The relationship with their study partner at the follow-up visit was also collected. Free ART was prescribed for HIV-1-infected partners who met the Chinese national treatment criteria (WHO stage 3 or 4 disease or CD4+ counts <200 cells/µL before 2007 or <350 cells/µL since 2008).17 Each participant was given compensation of ¥60 (∼US$9) for his/her time at each study visit.

Laboratory tests

HSV-2 infection was determined using the Captia anti-HSV-2 IgG ELISA (Trinity Biotech, Ireland). An index value >1.1 was classified as positive, and an index value of <0.90 was classified as negative. Equivocal samples (index value of 1.1–0.9) were retested; if it was still equivocal, it was classified as equivocal. Equivocal results may indicate true positives with low antibody levels, early HSV-2 infection or true negatives with a higher index value due to cross-reactivity with other infections (eg, herpes simplex virus type 1 (HSV-1), HIV-1).9 ,18 In addition, HIV-1 antibody, CD4+ count and HIV viral load were measured according to the manufacturers’ instructions, as previously described.15 ,16

Definition of HSV-2 acquisition outside the partnership

Incident HSV-2 infections among initially HSV-2 seronegative partners whose study partner was HSV-2 seronegative at baseline and did not seroconvert during follow-up, or whose study partner was HSV-2 equivocal at baseline but tested negative for HSV-2 at the follow-up visit were defined as acquisition outside the HIV-1-discordant partnership.

Statistical analysis

All statistical analyses were performed using SAS software (V.9.11, SAS Institute, USA). We compared categorical variables between those included and not included in the analysis using the χ2 test and Fisher's exact test when appropriate. For incidence and person-time calculations, it was assumed that HSV-2 seroconverted at the middle time point between the baseline and follow-up visit. Poisson 95% CIs were calculated for overall incidence and for subgroup comparisons. Univariate and multivariate cox proportional hazards regression models were used to examine factors independently associated with incident HSV-2 infections during follow-up, separately by HIV-1 serostatus at baseline. Variables with a p value less than 0.10 in univariate analyses or shown to be associated with HSV-2 acquisition in previous literature were included in multivariate analysis. The latter group of variables included demographic variables and the frequency of sex with study partner and outside sexual partnership during follow-up 9 ,19 Missing values were excluded from the corresponding analysis. We also conducted a subgroup analysis in which only initially HSV-2 negative participants were included and didn't find much difference in estimates and p values.

Results

Study population

A total of 1618 HIV-1-discordant couples were eligible for the study. One thousand one hundred and sixty-seven HIV-1-infected partners and 1052 HIV-1-uninfected partners completed the baseline survey and HSV-2 testing. Among them, 407 (34.9%) HIV-1-infected partners and 295 (28.0%) HIV-1-uninfected partners were HSV-2 positive, with highest HSV-2 prevalence among HIV-1-infected women (58.0%) and lowest HSV-2 prevalence among HIV-1-uninfected men (24.4%) (data not shown). Only 413 HIV-1-infected partners and 517 HIV-1-uninfected partners had a study partner who completed the baseline survey and HSV-2 testing were included in the final analysis (figure 1). The two groups (‘included in the analysis’ vs ‘not included in the analysis’) for HIV-1-infected and HIV-1-uninfected partners did not differ significantly with regard to most variables, such as sociodemographics, CD4+ counts and HSV-2 serostatus of their study partners. However, of those who were not included in the analysis, a greater proportion of HIV-1-infected and HIV-1-uninfected partners reported less frequent sex with study partner in the past 12 months (p<0.001 and p<0.001, respectively) and a greater proportion of HIV-1-infected partners had a plasma HIV viral load ≥1000 copies/mL and were not receiving ART at baseline (p<0.001) (data not shown).

Figure 1

Flow of study participants included for incidence analysis. HSV-2, herpes simplex virus type 2.

Baseline and follow-up characteristics

Among those included in the analysis, at baseline, similar proportions of HIV-1-infected and HIV-1-uninfected partners were HSV-2 equivocal (1.4%, 6/413 vs 1.5%, 8/517; p=0.906), but a lower proportion of HIV-1-infected partners had an initially HSV-2 positive partners than HIV-1-uninfected partners (11.4%, 47/413 vs 20.7%, 107/517; p=0.002). 207 (50.1%) HIV-1-infected partners were already receiving ART, 40.0% (151/377) had a CD4+ count <350 cells/μL and 40.3% (157/390) had HIV plasma viral load ≥1000 copies/mL.

At follow-up visit, similar proportions of HIV-1-infected and HIV-1-uninfected partners reported having sex with their study partner ≥5 times per month in the past 12 months (35.3%, 146/413 vs 33.1%, 171/517) and reported any unprotected sex with their study partner in the past 12 months (15.7%, 65/413 vs 16.1%, 83/517). A larger proportion of HIV-1-infected partners than HIV-1-uninfected partners reported any outside sexual partner during the follow-up period (8.7%, 36/413 vs 1.5%, 8/517). About 44 (1.1%) of HIV-1-infected partners initiated ART during follow-up.

HSV-2 and HIV-1 seroconversions

The median follow-up time for HIV-1-infected and HIV-1-uninfected partners was 17.7 months (IQR: 16.9–18.4) and 17.8 months (IQR: 16.9–18.4), respectively. HSV-2 seroconversions were observed in 49 persons (17 HIV-1-infected partners and 32 HIV-1-uninfected partners) during a total of 1301.8 person-years (PY), resulting in an overall incidence of 3.8 per 100 PY. HSV-2 incidence was 2.9 per 100 PY for HIV-1-infected partners and 4.5 per 100 PY for HIV-1-uninfected partners (crude HR (CHR)=0.65; 95% CI 0.36 to 1.17; p=0.157). Among those with an initially HSV-2 seronegative partner, HSV-2 incidence was 1.6 per 100 PY for HIV-1-infected partners and 1.9 per PY for HIV-1-uninfected partners (CHR=0.81; 95% CI 0.33 to 2.02; p=0.656). Among those with an initially HSV-2 positive partner, HSV-2 incidence was 12.7 per 100 PY for HIV-1-infected partners and 14.6 per PY for HIV-1-uninfected partners (CHR=0.86; 95% CI 0.38 to 1.94; p=0.709). Among those with an initially HSV-2 equivocal partner, HSV-2 incidence was 14.3 per 100 PY for HIV-1-infected partners and 10.0 per PY for HIV-1-uninfected partners (CHR=1.53; 95% CI 0.09 to 24.49; p=0.765). We also noted that 10 HIV-1-infected partners and 12 HIV-1-uninfected partners ended their relationship with study partner during follow-up, but there were no HSV-2 seroconversions among them.

In terms of sources of HSV-2 infections, according to the definition, 7 of 17 HSV-2 seroconversions among HIV-1-infected partners and 11 of 32 HSV-2 seroconversions among HIV-1-uninfected partners were classified as the acquisition outside the partnership. Taken together, at least 18 new HSV-2 infections were from an outside sexual partner, accounting for 36.7% (18/49) of the total infections.

HIV-1 seroconversions were observed among 7 of 517 HIV-1-uninfected partners during follow-up. Of the seven participants who acquired HIV-1 during follow-up, two were among 32 HSV-2 seroconverters (6.2%) and five were among 485 HSV-2 non-seroconverters (1.0%).

Risk factors for HSV-2 seroconversions in HIV-1-infected and HIV-1-uninfected partners

We examined risk factors for HSV-2 seroconversions separately among the HIV-1-infected and HIV-1-uninfected partners. Among HIV-1-infected partners, in multivariate analysis, having an initially equivocal HSV-2 result, having an initially HSV-2 positive or equivocal study partner who was HSV-2 seropositive, having no sex with study partner in the past 12 months (vs 1–4 times per month) and initiating ART during follow-up were significantly associated with increased risk of HSV-2 infection (table 1).

Table 1

Univariate and multivariate analyses of HSV-2 seroconversion risk factors among HIV-1-infected partners in HIV-1-discordant heterosexual couples (n=413)

Among HIV-1-uninfected partners, in multivariate analysis, having an initially HSV-2 seropositive study partner and having sex with study partner ≥5 times per month in the past 12 months (vs 1–4 times per month) were significantly associated with increased risk of HSV-2 infection, but having a study partner with baseline CD4+ count ≥350 cells/μL was significantly associated with decreased risk of HSV-2 infection (table 2).

Table 2

Univariate and multivariate analyses of HSV-2 seroconversion risk factors among HIV-1-uninfected partners in HIV-1-discordant heterosexual couples (n=517)

Discussion

To the best of our knowledge, this is one of the first cohort studies to investigate HSV-2 incidence and its associated risk factors, including the immunological status and ART initiation of HIV-1-infected partners among HIV-1-discordant heterosexual couples. HSV-2 incidence in our sample was lower than that recently reported among HIV-1-discordant couples in Kenya (14.8 per 100 PY).9 There are limited data available on HSV-2 incidence in China, mostly coming from female sex workers and men who have sex with men, with an estimated incidence of 21.9 per 100 PY and 8.1 per 100 PY, respectively.20 ,21 The relatively low overall incidence in our sample may be attributed to the lower proportion of participants with an initially HSV-2 seropositive partner, as well as the higher rates of condom use with study partner and relatively lower prevalence of high-risk sexual behaviour.

We found that those with an initially HSV-2 seropositive partner had the highest HSV-2 incidence during follow-up, which was consistent with a prior study.9 But we noted that at least 36% of incident HSV-2 infections were from an outside partner, slightly higher than the proportion of 30% observed in the other study of HIV-1-discordant couples from Kenya.9 In our study, incident HSV-2 infections among those having a study partner who was HSV-2 equivocal at baseline but tested negative for HSV-2 at the follow-up visit were counted as infection from outside the partnership, which was different from the Muiru et al9 study. Our results indicated that one or both members of HIV-1-discordant couples engaged in outside sexual partnerships. This suggests a potential risk of introducing HSV-2 and other STIs into the partnership and of transmitting HIV-1 to outside sexual partners.

Consistent with a prior study,9 we found that HIV-1-infected partners with an initially equivocal result were more likely to experience an HSV-2 seroconversion than those with an initially HSV-2 negative result and we also found that those having an initially equivocal partner were more likely to experience an HSV-2 seroconversion than those with an initially negative partner among HIV-1-uninfected partners. It is likely that equivocal results were either due to early HSV-2 infection or cross-reactivity with other viruses, including HSV-1 or HIV-1.9 ,18 Therefore, caution in the use and interpretation of the equivocal results is advised; a confirmatory assay (western blot) is advised to be performed for equivocal results.22

We observed that among HIV-1-uninfected partners, the frequency of sex with their study partners was associated with increased risk of HSV-2 seroconversion, which is consistent with the findings of a previous study.23 It is of note that among HIV-1-infected partners, there was a higher risk of HSV-2 seroconversion among those who reported no sex with their study partners. The possible explanation is that they instead engaged in outside sexual partnership, which put them at risk of HSV-2 infection.

A notable finding was that HIV-1-infected partners who initiated ART during follow-up were at a higher risk of HSV-2 acquisition. Previous studies observed an increased frequency of HSV-2 shedding and genital ulcer diseases after ART initiation due to immune reconstitution among HIV-1-infected patients,13 ,14 ,24 so ART initiation could increase the risk of transmitting HSV-2 from HIV/HIV-2 co-infected persons to their sexual partners. Ours was the first study to demonstrate that ART initiation is associated with HSV-2 acquisition. It is possible that ART initiation may increase sexual risk taking, 25 ,26 which increases the risk of HSV-2 infection. However, we didn't find an increase of having sex with outside partners or having more sex with study partner in our sample, suggesting that this may not fully explain the finding. As ART was initiated usually based on CD4+ count, the low CD4+ counts or poor immune competence may increase HIV-1-infected partners’ susceptibility to HSV-2 infection, but requires further investigation.

Our findings provide new evidence for the relationship between HIV-1 and HSV-2 infections. We found that HIV-1-uninfected partners whose HIV-1-infected partner had CD4+ count ≥350 cells/μL at baseline were at decreased risk of HSV-2 infection than those whose HIV-1-infected partner had CD4+ count <350 cells/μL. The possible biological mechanism was that HSV-2 viral shedding increased as the CD4+ counts decreased.10–12

Our study has several limitations. First, all participant couples were recruited from a rural area in China; these findings may not be generalisable to other areas. Second, a substantial number of subjects were lost to follow-up, and thus our findings were subject to selection bias; however, we found no significant differences in most baseline variables between those remained in follow-up and those lost to follow-up, suggesting that the risk estimates should not be severely affected. Third, the frequency of unprotected sex and outside sexual partners may be underestimated due to social desirability bias. In addition, HSV-2 testing was only taken at baseline and at follow-up visit, and HSV-2 seroconversion was presumed to have occurred between them. The interval censoring may result in biased parameter estimates (bias direction unknown) and SEs which are biased downwards.27

In conclusion, we found that HSV-2 incidence was high among those with an initially HSV-2 positive partner, underscoring the importance of developing prevention and intervention programmes to prevent HSV-2 transmission among this population. Specifically, routine HSV-2 screening and education on correct and consistent condom use should be included in HIV prevention programmes. HSV-2 suppressive therapy with acyclovir is also recommended, which has been shown to reduce HSV-2 transmission among HIV-1-discordant couples.28 Furthermore, the high proportion of HSV-2 due to extramarital sex highlights the need for additional efforts to address this problem. The relationship between ART initiation and HSV-2 seroconversion requires further investigation.

Key messages

  • At least 36% of incident HSV-2 infections were from outside sexual partnership.

  • HIV-1-infected partners who initiated antiretroviral therapy during follow-up were at a higher risk of HSV-2 acquisition.

  • HSV-2 incidence among HIV-1-uninfected partners was associated with a lower CD4+ count of their HIV-1-infected partners at baseline.

Acknowledgments

We would like to thank all the participants.

References

Supplementary materials

  • Abstract in Chinese

    This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.

Footnotes

  • Handling editor Jackie A Cassell

  • Contributors NH, SD and ZW designed and/or supervised the study. YD, ZW, SD, KR, YY, JW, MG, RY and NH implemented the study. JW and GM performed the laboratory tests. YD performed the statistical analyses. YD and NH wrote the first draft of manuscript. RD reviewed and provided critical comments on manuscript drafts. All authors were involved in the interpretation of the results, provided feedback on several versions of the manuscript and approved the final manuscript.

  • Funding This study was supported by the National Science and Technology Major Project on Prevention and Treatment of Major Infectious Diseases, including AIDS and Viral Hepatitis from the Chinese Ministry of Health (Grant No. 2008ZX10001-016), and the Natural Science Foundation of China (Grant No. 81072345).

  • Competing interests None declared.

  • Ethics approval This study was approved by the institutional review board of the Chinese National Center for AIDS/STD Control and Prevention, Chinese Center for Disease Control and Prevention.

  • Provenance and peer review Not commissioned; externally peer reviewed.