Sage Journals: Discover world-class research

Abstract

Background

For women living with HIV (WLHIV), co-infection with herpes simplex virus type 2 (HSV-2) causes severe genital ulcers and presents additional challenges for their HIV care. To inform preventive strategies, we aimed to determine the incidence and risk factors of HSV-2 positivity in a prospective cohort of South African women.

Methods

The CAPRISA 002 study enrolled women at acute HIV infection between 2004 and 2020. HSV-2 testing was conducted by multiplex polymerase chain reaction (PCR) assay on collected vaginal swabs up to twice annually during follow-up. We calculated incidence as the number of new cases per 100 person-years (PYs) and used Cox-proportional-hazard regression to identify factors associated with time-to-HSV-2 PCR positivity.

Results

At enrolment, the median age of 171 women was 24 years, interquartile range (IQR 21–28), and the estimated median days since HIV infection was 42 (IQR 22–65). Of participants tested at enrolment, HSV-2 antibody prevalence was 81.4% (105/129), and 10.6% (12/113) were positive by PCR. Among 147 women with a prior negative HSV-2 PCR diagnosis, we observed 47 new HSV-2 PCR positive cases over 424.4 PYs of follow-up, yielding an incidence rate of 11.1 cases per-100-PYs. HSV-2 PCR positivity incidence was higher among younger women (<25 years: adjusted Hazard Ratio [aHR] = 5.91, 95%CI 3.02–11.6), those with bacterial vaginosis (BV) (Nugent score 7–10: aHR = 2.17, 95%CI 1.15–4.10) and lower CD4 counts (<500 cells/μl: aHR = 2.04, 95%CI 1.08–3.87).

Conclusion

After acute HIV infection in women, the incidence of HSV-2 PCR positivity was associated with younger age, BV diagnosis and lower CD4 count.

Keywords

Sexually transmitted infections human immunodeficiency virus women living with human immunodeficiency virus herpes simplex virus type 2 incidence

Introduction

Sexually transmitted infections (STIs) facilitate Human Immunodeficiency Virus (HIV) infection and transmission.¹ Herpes simplex virus (HSV) type 2 is among the leading risk factors for HIV infection mediated through genital ulcers or herpes disease.² HSV type 1 also causes genital herpes, but recurrent disease and subclinical viral shedding are more frequent with HSV-2, which affects more women than men.^3,4 Comparing global estimates, the highest contribution of HSV-2 infection to population-level HIV incidence comes from the World Health Organisation (WHO) Africa Region, with an estimated population attributable fraction of 37.1% compared to 11.6 to 21.3% from the rest of the world.⁵

Among women living with HIV (WLHIV), the risk of HSV-2 infection^5,6 and HSV-2 shedding at the vaginal mucosa⁷ are higher than in HIV-negative women. In South Africa, a global hotspot of the HIV epidemic,⁸ two studies have estimated 13.7% prevalence of HSV-2 in men and women living with HIV⁹ and 10.8% prevalence in WLHIV.¹⁰ In WLHIV, HSV-2 coinfection with HIV causes increased morbidity through prolonged genital herpes recurrences, especially among patients with lower immune status.³ This makes their HIV care more burdensome due to more clinic visits and treatments, often leading to poorer clinical outcomes.¹¹

Considering the chronicity of HSV-2 infection and the current non-availability of an effective cure, prevention is preferable, especially among WLHIV to reduce their healthcare burden and risk of HIV transmission, but no effective vaccine is currently available.³ To strengthen the evidence on HSV-2 occurrence and inform public health prevention efforts, this study aimed to determine HSV-2 positivity incidence and the associated demographic, sexual-behavioural and clinical risk factors among women enrolled at acute HIV infection and followed up before, after, and during antiretroviral treatment (ART) in South Africa.

Methods

Study design, population, and setting

This retrospective cohort study used data from the Centre for the AIDS Programme of Research in South Africa (CAPRISA) acute HIV infection study (CAPRISA 002) conducted from 2004 to 2020 to investigate the disease progression of HIV-1 subtype C infection among South African adult women.¹² Participants were enrolled from other CAPRISA HIV prevention studies upon HIV seroconversion.¹² The eligibility criteria for the study included being 16 years or older, having a haemoglobin >9.0 g/dl and a negative HIV antibody test result within 5 months of HIV infection. The study confirmed HIV infection at enrolment with Enzyme Immunoassay (EIA) antibody testing, and HIV reverse transcription polymerase chain reaction (RT-PCR) testing was conducted when the antibody result was negative. The time of HIV infection was determined as the midpoint date between the last documented negative and the first positive HIV antibody test. For participants with a negative antibody test at enrolment, the estimated HIV infection date was 14 days before the positive RT-PCR test. Testing and management of HSV-2 infection and other STIs were conducted bi-annually before and annually for up to 5 years after ART initiation, with an additional 15 years of optional follow-up offered thereafter. Participants started ART based on the applicable South African guidelines at the time. HIV pre-and post-test counselling, risk reduction counselling, male and female condoms, access to clinical care, and treatment for STIs were provided to all participants throughout the study. The Biomedical Research Ethics Committee (BREC) of the University of KwaZulu-Natal approved the CAPRISA 002 study (E013/04) and this analysis (BREC/00003808/2022).

Primary outcome

The primary outcome was HSV-2 positivity incidence based on PCR testing. HSV-2 positivity incidence was defined as any new diagnosis after an initial negative test at enrolment.

Exposure variables

Exposure variables included demographic characteristics namely age and educational status, sexual behavioural characteristics namely age at first sexual intercourse, number of sexual partners and condom use and clinical characteristics including viral load, CD4 count, BV and STIs.

Laboratory measures

For HSV-2 PCR testing on vaginal swabs we used an in-house multiplex PCR assay at the National Institute for Communicable Diseases (NICD) laboratories in Johannesburg, South Africa. Blood specimens were collected and sent to Global laboratories in Durban for HSV-2 antibody testing based on HSV-2 glycoproteins G-2 with the HerpeSelect-2 EIA (Focus Technologies, Cypress, California, USA). The FACSCalibur flow cytometer (BD Biosciences, San Jose, California) was used to measure CD4+ T cell count. ROCHE COBAS Ampliprep-COBAS TaqMan version 2.0 and COBAS Amplicor version 1.5 RT-PCR and assays (Roche Diagnostics, Rotkreuz, Switzerland) were used to measure HIV viral loads. Urine specimens and vaginal swabs were collected and tested for Neisseria gonorrhoeae (NG), Chlamydia trachomatis (CT), Trichomonas vaginalis (TV) and Mycoplasma genitalium (MG), with either laboratory-based (in-house multiplex assay) or point-of-care (GeneXpert) assays. Gram-stained vaginal swabs were assessed with a light microscope by a certified laboratory technician using Nugent’s criteria for diagnosing BV (scores 7–10).

Data collection

Data capture for demographic information, physical examination, sexual behavioural risk assessments and clinical measures were conducted using structured case report forms (CRFs). Trained research assistants completed the CRFs at the study clinics and faxed them to the CAPRISA data management system using the iDataFax system (DF/Net Research, Seattle, USA). All data entry activities followed the CAPRISA data management Standard Operating Procedures for quality control.

Statistical analyses

We performed all statistical analyses and plotted graphs with R 4.2.1 (R Foundation for Statistical Computing, Vienna, Austria).¹³ We summarized participant characteristics at baseline with proportions and medians stratified by HSV-2 PCR testing status (‘Negative’, ‘Positive’, ‘Not tested’) since not all participants were tested at enrolment. We compared the baseline differences in proportions and medians with the Chi-squared and Kruskal-Wallis tests.

All subjects who were HSV-2 PCR negative at their first time of testing, either at enrolment or subsequently for those not tested at enrolment, were included in the incidence analyses. We conducted a single-failure survival analysis to determine the incidence of HSV-2 PCR positivity measured as the number of new cases per 100 person-years (PYs) follow-up. The time at risk started from the first HSV-2 PCR negative to the first positive date or censoring. HSV-2 PCR was assumed to occur on the testing date. We then fitted Cox-proportional-hazard regression models using Breslow’s method for tied survival times to determine the risk factors associated with the time to HSV-2 PCR positivity adjusted for age, number of lifetime casual sex partners, condom use for contraception, co-occurrence of BV, CD4 count and HIV viral load status. We evaluated the multivariable Cox regression model for proportionality of hazard rates over time with the Schoenfeld's residuals test.

To provide a graphical representation of the probability of HSV-2 PCR positivity over time, based on the multivariable Cox model, we fitted adjusted Cox failure curves for each statistically significant variable. These were done by plotting the HSV-2 PCR failure probabilities and observation time obtained from separate fitted Cox models where each statistically significant variable was specified as the strata and adjusted for the other covariates. On the Y axis of the Cox failure curves, we plotted the adjusted probability of HSV-2 PCR positivity and plotted the person-time or years of follow-up after acute HIV infection on the X axis. We used the log-rank test to determine the equality of failure curves over time and a 5% significance level for all hypothesis tests.

Results

Of the women enrolled at acute HIV infection in the CAPRISA 002 study, 171 had at least one HSV-2 PCR test performed during follow-up between October 2004 and November 2020 and were included in this study. Among the 171 women, the median follow-up time was 9.2 years (interquartile range [IQR] 6.2–11.5) with a median of 6 (IQR 4–8) visits, where HSV-2 PCR testing was conducted. During follow-up, 136/171 women (79.5%, 95% Confidence Interval [CI] 72.7–85.3) started ART at a median of 4.8 (IQR 2.6–6.5) years after enrolment.

Characteristics of participants at enrolment

The baseline characteristics of the women are shown in Table 1, stratified by HSV-2 PCR testing status. Participants were young women (median age 24 years, IQR 21–28), and most were diagnosed with HIV infection within 2 months of HIV acquisition (median 42 days IQR 22–65). About half of the women (46.2%) had completed high school, and 43.9% reported their first sexual experience before 18 years. Furthermore, 29.5% reported 2 or more lifetime sexual partners, and 15.9% reported a recent casual sexual partner. Condom use for pregnancy prevention was reported by 39.3% of the women. The median CD4 count at enrolment was 508 cells/μl (IQR 389–671), and the median HIV log₁₀ viral load was 4.6 copies/mL (IQR 3.9–5.3). Of the women tested at enrolment, the prevalence of HSV-2 PCR was 10.6%, and 81.4% were HSV-2 antibody positive. Other STIs were also common at enrolment; 8.6% had NG, 19.7% had CT, 14.5% had TV, 8.6% had MG, and 61.5% had BV.

Table 1.

Characteristics of participants at enrolment stratified by HSV-2 PCR testing status.

Variable	Category	Overall (n = 171) % (n)	Negative (n = 101) % (n)	Positive (n = 12) % (n)	Not tested (n = 58) % (n)	^ap-value
Age in years	Median (IQR)	24 (21–28)	24 (21–28)	24.5 (21.5–26.5)	25 (21–30)	0.724
Age-group (years)	18–24	52.6 (90/171)	55.5 (56/101)	50.0 (6/12)	48.3 (28/58)	0.356
	25–34	36.3 (62/171)	37.6 (38/101)	33.3 (4/12)	34.5 (20/58)
	35+	11.1 (19/171)	6.9 (7/101)	16.7 (2/12)	17.2 (10/58)
School level completed (high)	1st year of high school	7.0 (12/171)	4.0 (4/101)	0.0 (0/12)	13.8 (8/58)	0.090
	3^rd year of high school	46.8 (80/171)	49.5 (50/101)	33.3 (4/12)	44.8 (26/58)
	High school	46.2 (79/171)	46.5 (47/101)	66.7 (8/12)	41.4 (24/58)
Age at first sexual debut (years)	Median (IQR)	18 (16–19)	18 (16–19)	18 (16–20)	18 (16–19)	0.862
Age at first sexual debut (group)	<18 years	43.9 (75/171)	45.5 (46/101)	41.7 (5/12)	41.4 (24/58)	0.867
Age at first sexual debut (group)	>=18 years	56.1 (96/171)	54.5 (55/101)	58.3 (7/12)	58.6 (34/58)
^bCasual sexual partners in lifetime	No partner	64.5 (107/166)	71.4 (70/98)	75.0 (9/12)	50.0 (28/56)	0.006
	One partner	6.0 (10/166)	6.1 (6/98)	16.7 (2/12)	3.6 (2/56)
	2+ partners	29.5 (49/166)	22.5 (22/98)	8.3 (1/12)	46.4 (26/56)
^bCasual sexual partners in recent 3 months	No partner	77.1 (131/170)	84.0 (84/100)	91.7 (11/12)	62.1 (36/58)	0.003
	One partner	7.1 (12/170)	7.0 (7/100)	8.3 (1/12)	6.9 (4/58)
	2+ partners	15.9 (27/170)	9.0 (9/100)	0.0 (0/12)	31.0 (18/58)
Contraceptive condom-use	Yes	39.3 (66/168)	34.3 (34/99)	25.0 (3/12)	50.9 (29/57)	0.072
Days since HIV infection	Median (IQR)	42 (22–65)	45 (25–67)	39.5 (22.5–62.5)	34.5 (16–62)	0.195
BMI (kg/m²)	<18.5	2.4 (4/169)	2.0 (2/101)	8.3 (1/12)	1.8 (1/56)	0.398
	18.5–24.9	38.5 (65/169)	36.6 (37/101)	41.7 (5/12)	41.1 (23/56)
	25.0–29.9	26.0 (44/169)	24.6 (25/101)	8.3 (1/12)	32.1 (18/56)
	>30	33.1 (56/169)	36.6 (37/101)	41.7 (5/12)	25.0 (14/56)
CD4 count (cells/μl)	Median (IQR)	508 (389–671)	510 (393–643)	493 (395–682.5)	506 (377–771)	0.869
CD4 count (cells/μl)	<200	2.1 (3/146)	3.2 (3/95)	0.0 (0/12)	0.0 (0/39)	0.686
	200–499	46.6 (68/146)	45.3 (43/95)	58.3 (7/12)	46.2 (18/39)
	500+	51.4 (75/146)	51.6 (49/95)	41.7 (5/12)	53.9 (21/39)
HIV viral load (copies/mL)	Detectable status	95.9 (163/170)	96.0 (97/101)	91.7 (11/12)	96.5 (55/57)	0.741
Log₁₀ HIV viral load (copies/mL)	Median (IQR)	4.6 (3.9–5.3)	4.6 (4.0–5.3)	3.9 (3.3–4.3)	4.9 (4.0–5.5)	0.021
HSV-2 antibodies	Positive	81.4 (105/129)	83.3 (75/90)	100.0 (11/11)	67.9 (19/28)	0.047
HSV-2 PCR	Positive	10.6 (12/113)	—	—	—	—
Bacterial vaginosis	Positive	61.5 (72/117)	65.0 (65/100)	50.0 (6/12)	20.0 (1/5)	^c0.086
Neisseria gonorrhoeae	Positive	8.6 (10/117)	8.9 (9/101)	8.3 (1/12)	0.0 (0/4)	^c0.822
Chlamydia trachomatis	Positive	19.7 (23/117)	20.8 (21/101)	0.0 (0/12)	50.0 (2/4)	^c0.053
Trichomonas vaginalis	Positive	14.5 (17/117)	15.8 (16/101)	0.0 (0/12)	25.0 (1/4)	^c0.221
Mycoplasma genitalium	Positive	8.6 (10/117)	8.9 (9/101)	8.3 (1/12)	0.0 (0/4)	^c1.000

HSV = Herpes simplex virus, PCR = Polymerase Chain Reaction, HIV = Human Immunodeficiency Virus, WLHIV = Women Living with HIV, Kg = Kilograms, BMI = Body Mass Index, μl = Microlitre, mL = Millilitre, IQR = Inter Quartile Range, CI = Confidence Interval.

Denominators that do not equal the sample sizes are due to missing data. Percentages may not add up to 100 because of rounding.

^ap-value compares (Negative, Positive, and Not tested).

^bCasual partners in the last 3 months exclude transaction sex among sex workers in the sample.

^cFisher’s exact test.

Incidence of HSV-2 PCR positivity after acute HIV infection among women

Of the 171 women tested during follow-up in the CAPRISA 002 study, 155 (90.6%) were HSV-2 PCR negative at their first testing episode. Of these 155 women, 8 were lost to follow-up, and the remaining 147 were included in the incidence analysis. Over a total of 424.4 PYs, 47 new HSV-2 PCR positive cases were observed, at a cumulative incidence of 32.0% (95% CI 24.5–40.2) and a crude incidence of 11.1 cases per 100 PYs (95% CI 8.3–14.6).

Factors associated with time to HSV-2 PCR positivity after acute HIV infection among women

The factors associated with time to HSV-2 PCR positivity in the Cox regression analyses are presented in Table 2. After adjusting for other covariates, a positive HSV-2 PCR test occurred sooner among younger women (<25 years: adjusted Hazard Ratio [aHR] 5.91, 95% CI 3.02–11.60, p < .001), those with BV (aHR 2.17, 95% CI 1.15–4.1, p = .017) and those with lower CD4 counts (<500 cells/μl: aHR 2.04, 95% CI 1.08–3.87, p = .028). There was a weak association (protective effect) between condom use and HSV-2 PCR incidence (aHR 0.54, 95% CI 0.28–1.03, p = .060).

Table 2.

Cox regression for factors associated with time to HSV-2 PCR positivity among WLHIV.

Variable	Category	n	Person-years	No. of cases	HR (95%CI)	p-value	aHR (95% CI)	p-value
^aAge in years	>= 25	104	349.99	25	1	—	1	—
^aAge in years	<25	43	74.38	22	4.77 (2.58-8.84)	<0.001	5.91 (3.02-11.6)	<0.001
^bLifetime casual sexual partners	No partner	89	236.29	28	1	—	1	—
	One partner	8	24.30	1	0.36 (0.05-2.62)	0.310	0.90 (0.11-7.27)	0.924
	2+ partners	46	155.56	17	0.94 (0.51-1.73)	0.836	1.46 (0.75-2.87)	0.267
Condom use for contraception	No	89	249.58	31	1	—	1	—
Condom use for contraception	Yes	56	174.23	16	0.75 (0.41-1.37)	0.344	0.54 (0.28-1.03)	0.060
^aBV	Negative	82	245.80	18	1	—	1	—
^aBV	Positive	65	178.57	29	2.30 (1.26-4.18)	0.007	2.17 (1.15-4.1)	0.017
^aCD4 count (cells/μl)	>= 500	67	217.20	18	1	—	1	—
^aCD4 count (cells/μl)	<500	80	207.17	29	1.71 (0.95-3.09)	0.074	2.04 (1.08-3.87)	0.028
^aHIV viral load	Undetectable	22	85.34	5	1	—	1	—
^aHIV viral load	Detectable	125	339.03	42	2.20 (0.84-5.73)	0.107	1.31 (0.47-3.68)	0.609

HSV = Herpes simplex virus, HIV = Human Immunodeficiency Virus, WLHIV = Women Living with HIV, μl = Microlitre, CI = Confidence Interval, HR = Hazard Ratio, aHR = Adjusted Hazard Ratio.

^aVariable measured on the day of HSV-2 PCR incidence or last study visit for censored observations (all other variables were measured only at enrolment).

^bCasual partners in the last 3 months exclude transaction sex among sex workers in the sample.

Cox failure curves showing the adjusted probability of HSV-2 PCR positivity over time

Figure 1 displays trends of the adjusted probabilities of HSV-2 PCR positivity by age group, BV and CD4 count in panels A, B and C. The graphs represent estimates from the multivariable Cox regression analysis, as explained in the Methods section. The Y axis of the graph shows the adjusted probability of HSV-2 PCR positivity, and the X axis shows the years of follow-up after acute HV infection. The curves show a significantly faster trend in HSV-2 PCR positivity by (A) younger age of < 25 versus >= 25 years, (B) BV positive versus negative status and (C) lower CD4 count of < 500 versus >= 500 cells/μl. The log-rank test p-values for all failure curves were less than 0.05.

Figure 1.

Adjusted Cox failure curves showing trends of the probability of HSV-2 PCR positivity by (a) Age-group, (b) Bacterial vaginosis status and (c) CD4 count (cells/microliter). All curves were adjusted for other variables in the multivariable Cox regression: Age-group, Lifetime and casual sexual partners, BV status, condom use for contraception, CD4 count and HIV viral load.

Multivariable Cox regression model goodness of fit: the proportional-hazards assumption

Figure 2 shows a graphical representation of results from the multivariable Cox regression model goodness of fit test. It displays plots of the scaled Schoenfeld’s residuals against observation time in years for each variable in the multivariable Cox regression. The figure also shows the p-values of the Schoenfeld test of the proportional hazard assumption for each covariate and the global p-value for the multivariable model. The Schoenfeld residual is the difference between each covariate’s observed and expected hazard ratios at the time of HSV-2 PCR occurrence. A stable trend of the residuals against time indicates the proportionality of the HSV-2 PCR hazard rates over time.¹⁴ In each graph, the black dotted points represent the residuals versus time plot, the solid black lines represent the smoothing spline fit to the plot, and the black dashed lines represent the 95% confidence interval bands. All plots display a stable trend of the residuals over time, indicating that the model meets the proportional hazard assumption (all covariates p > .05 and global p = .903).

Figure 2.

Plots of the scaled Schoenfeld’s residuals against observation time for all covariates in the multivariable Cox regression model. The black dotted points represent the residuals versus time plot. The solid black lines represent the smoothing spline fit to the plot. The black dashed lines represent the 95% confidence interval bands.

Discussion

This study investigated the incidence and risk factors associated with HSV-2 PCR positivity from acute to chronic HIV infection in a cohort of women from high-risk communities in South Africa. Although the crude incidence of HSV-2 PCR positivity was 11.1 cases per 100 PYs among women with an initial negative test, the incidence was higher among younger women, those with BV and lower CD4 counts after adjusting for other participant characteristics in the multivariable Cox regression analysis. Of note, the incidence analysis included only women with a prior negative HSV-2 PCR diagnosis, but of the 171 women included in this study, 81.4% (n = 105/129) of those tested at enrolment were HSV-2-antibody-positive indicating a prior HSV-2 infection. Thus, our HSV-2 PCR positivity outcome predominantly represents HSV-2 reactivation or viral shedding from existing disease with occasional instances of newly acquired infection. Studies have shown that HSV-2 viral shedding or re-activation in WLHIV is associated with increased HIV viral shedding^15,16 which can further increase HIV infectiousness.^2,5 Aligned with these studies, our findings emphasize the importance of reducing the risk of HSV-2 incidence, viral shedding, or re-activation in WLHIV to decrease population level HIV transmission.

In sub-Saharan Africa, a moderate number of studies^17–19 have reported a high HSV-2 seroprevalence among WLHIV, with an estimated 93.3% prevalence reported in KwaZulu-Natal, South Africa¹⁹ where this study was conducted. We found two studies in South Africa that performed diagnostic testing to determine active HSV-2 infection prevalence of 10.8% in WLHIV¹⁰ and 13.7%⁹ in PLHIV. However, few studies have determined the incidence and predictors of HSV-2 PCR positivity among PLHIV in sub-Saharan Africa. One study in Uganda enrolled 924 participants, including 45 (4.9%) PLHIV, to determine the seroprevalence and incidence of genital ulcer infections, including HSV-2.²⁰ Of the 45 PLHIV who were initially HSV-2 seronegative, 5 seroconverted based on serological testing during 15.0 person-years of observation at an incidence rate of 333.3 cases per 1000 PYs.²⁰ Among 343 young PLHIV included in the Reaching for Excellence in Adolescent Care and Health (REACH) study in the United States, HSV-2 incidence was 8.5 cases per 100 person-years based on antibody testing in participants who were seronegative at baseline.²¹

We found an association between younger age and the incidence of HSV-2 PCR positivity, an indication of higher risk of HSV-2 infection or reactivation^3,22 among this age group. This is consistent with established evidence of higher STI infection risk in younger women²³ as previously demonstrated in this same cohort of WLHIV for other STIs, including NG, CT, TV or MG.²⁴ Several reasons have been reported for the high STI burden in younger women including riskier sexual practices like multiple partners and inconsistent condom use,^1,25 often exacerbated by gender and age-related power imbalances in sexual relationships.²⁶

HSV-2 PCR positivity incidence was also associated with BV diagnosis in our study. BV is known to be related to genital inflammation through the production of inflammatory cytokines²⁷ in the vaginal mucosa, which further increases the risk of new HSV-2 infections and HSV-2 viral shedding²⁸ in women with pre-existing subclinical disease. Moreover, women with HSV-2 infection usually have a higher occurrence of BV,^29,30 as HSV-2 also causes genital inflammation, generally resulting in an increased presence of BV-causing bacteria.³¹ Findings from these studies and ours, thus, confirm a bidirectional association between HSV-2 infection and BV occurrence in women,³² an interaction increasing the risk for opportunistic infections and worsening reproductive and sexual health in women.

We found an association between lower CD4 count and the incidence of HSV-2 PCR positivity among WLHIV. Consistent with the relationship between immune status and other opportunistic infections, our finding emphasises the potential benefit of immune reconstitution through ART in reducing HSV-2 co-infection and related clinical consequences in WLHIV. Consistent with our finding, a cohort study conducted in Burkina Faso revealed that staying longer on ART was associated with reduced odds of HSV-2 viral shedding and genital ulcer disease, mainly in participants who achieved HIV viral suppression.¹⁸ After introducing the universal ‘Test and Treat’ policy for HIV care in South Africa, a large-scale programmatic data analysis has shown better treatment outcomes with early ART initiation at CD4 count >500 cells/μ.³³ Our findings, thus, underscore the need to sustain such interventions with high treatment adherence for improved outcomes and prevention of opportunistic co-morbidities among PLHIV.

Our study had some limitations. Firstly, the sexual-behavioural characteristics, like condom use and number of sexual partners, were self-reported. Thus the related results could have been prone to social desirability³⁴ and recall biases.³⁵ Second, our study cohort included women recruited from high-risk communities for STI/HIV and may therefore not represent the general population of South Africa. For these reasons, we included variables on sexual partners in our multivariable regression models. Third, HSV-2 seroprevalence was high in this cohort at the time of enrolment. We were, therefore, unable to determine whether all new HSV-2 PCR positive cases were truly new infections or represented intermittent shedding of HSV-2.

In conclusion, our study has demonstrated a higher risk of HSV-2 infection or reactivation in younger WLHIV. It is important to promote safe sexual practices such as condom use through evidence-based behavioural interventions during chronic HIV infection with a special focus on younger women. The findings also support the importance of immune reconstitution with ART for improving not only HIV-treatment outcomes but also reducing HSV-2 co-infection or viral shedding and, as a result, reduce the risk of HIV transmission. Thus, it is necessary to improve HIV treatment continuation, retention-in-care and adherence through enhanced adherence counselling particularly in participants with poorer HIV treatment outcomes during ART. Our findings also demonstrate the role of genital inflammation through BV diagnosis for HSV-2 risk. This underscores the need for comprehensive STI/BV care, including HSV-2 testing and management as part of routine HIV care, to control STIs in general and reduce HSV-2 incidence and reactivation or shedding in WLHIV, specifically. Put together, adherence to these recommendations can improve the quality of life of WLHIV and reduce their HIV infectiousness which can result in lower population level HIV transmission.

Footnotes

Acknowledgements

We are grateful to all CAPRISA 002 study participants for participating in this research and the study team for conducting the study.

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: Funding for the CAPRISA 002 Acute HIV Infection Study was received from the Comprehensive International Program of Research on AIDS (CIPRA), funded by the National Institute of Allergy and Infectious Disease (NIAID), National Institutes of Health (NIH) and the US Department of Health and Human Services (DHHS) (Grant#1 U19 AI51794). KA was supported by HEARD scholarship with funding provided by Swedish International Development Cooperation (SIDA). The funders played no role in the study design, data collection and analysis, publication decision, or manuscript preparation.

Ethical statement

Contributors

KA, AT, and NG designed the study and are accountable for the overall content of the manuscript. KA conducted the statistical analysis and prepared the manuscript. All authors critically reviewed and approved the final submitted version of the manuscript.

ORCID iD

Kwabena Asare

References

Centres for Disease Control and Prevention . STDs and HIV – CDC detailed fact sheet, https://www.cdc.gov/std/hiv/stdfact-std-hiv-detailed.htm#foot1 (2022, accessed 11 April 2022).

Bradley

Floyd

Piwowar-Manning

, et al. Sexually transmitted bedfellows: exquisite association between HIV and herpes simplex virus type 2 in 21 communities in Southern Africa in the HIV prevention trials network 071 (PopART) study. J Infect Dis 2018; 218: 443–452. DOI: 10.1093/infdis/jiy178.

Centers for Disease Control and Prevention . Genital herpes – CDC detailed fact sheet, https://www.cdc.gov/std/herpes/stdfact-herpes-detailed.htm (2022, accessed 14 October 2022).

Akinyi

Odhiambo

Otieno

, et al. Prevalence, incidence and correlates of HSV-2 infection in an HIV incidence adolescent and adult cohort study in western Kenya. PLoS One 2017; 12: e0178907. DOI: 10.1371/journal.pone.0178907.

Looker

Welton

Sabin

, et al. Global and regional estimates of the contribution of herpes simplex virus type 2 infection to HIV incidence: a population attributable fraction analysis using published epidemiological data. Lancet Infect Dis 2020; 20: 240–249. DOI: 10.1016/s1473-3099(19)30470-0.

Rajagopal

Magaret

Mugo

, et al. Incidence of herpes simplex virus type 2 Infections in Africa: a systematic review. Open Forum Infect Dis 2014; 1: ofu043–ofu043. DOI: 10.1093/ofid/ofu043.

Augenbraun

Feldman

Chirgwin

, et al. Increased genital shedding of herpes simplex virus type 2 in HIV-seropositive women. Ann Intern Med 1995; 123: 845–847. DOI: 10.7326/0003-4819-123-11-199512010-00006.

Wand

Ramjee

. Targeting the hotspots: investigating spatial and demographic variations in HIV infection in small communities in South Africa. J Int AIDS Soc 2010; 13: 41–41. DOI: 10.1186/1758-2652-13-41.

Onoya

Manjengwa

Nattey

, et al. Incidence and predictors of sexually transmitted infections among adult HIV-positive patients receiving antiretroviral therapy at Themba Lethu HIV clinic in Johannesburg, South Africa. S Afr Med J 2021; 111(1): 62–62.

10.

Samie

Mnisi

Ramantswana

. Prevalence of HSV1 and HSV2 among HIV patients in northern South Africa as determined by real time PCR in urine samples. Am J Infect Dis 2017; 13: 3–12. DOI: 10.3844/ajidsp.2017.3.12.

11.

Pourcher

Gourmelen

Bureau

, et al. Comorbidities in people living with HIV: an epidemiologic and economic analysis using a claims database in France. PLoS One 2020; 15: e0243529–20201217. DOI: 10.1371/journal.pone.0243529.

12.

Van Loggerenberg

Mlisana

Williamson

, et al. Establishing a cohort at high risk of HIV infection in South Africa: challenges and experiences of the CAPRISA 002 acute infection study. PLoS One 2008; 3: e1954. DOI: 10.1371/journal.pone.0001954.

13.

R Core Team . R. A language and environment for statistical computing, 2022, https://www.R-project.org/.

14.

Box-Steffensmeier

Jones

. Event history modeling: a guide for social scientists. Cambridge: Cambridge University Press, 2004.

15.

McClelland

Wang

Overbaugh

, et al. Association between cervical shedding of herpes simplex virus and HIV-1. Aids 2002; 16: 2425–2430. DOI: 10.1097/00002030-200212060-00007.

16.

Stinn

Kuntz

Varon

, et al. Subclinical genital herpes shedding in HIV/herpes simplex virus 2-coinfected women during antiretroviral therapy is associated with an increase in HIV tissue reservoirs and potentially promotes HIV evolution. J Virol 2020; 95: 20201209. DOI: 10.1128/jvi.01606-20.

17.

Schaftenaar

Verjans

GMGM

Getu

, et al. High seroprevalence of human herpesviruses in HIV-infected individuals attending primary healthcare facilities in rural South Africa. PLoS One 2014; 9: e99243. DOI: 10.1371/journal.pone.0099243.

18.

Low

Nagot

Weiss

, et al. Herpes simplex virus type-2 cervicovaginal shedding among women living with HIV-1 and Receiving antiretroviral therapy in burkina faso: an 8-year longitudinal study. J Infect Dis 2016; 213: 731–737. DOI: 10.1093/infdis/jiv495.

19.

Kharsany

ABM

McKinnon

Lewis

, et al. Population prevalence of sexually transmitted infections in a high HIV burden district in KwaZulu-Natal, South Africa: implications for HIV epidemic control. Int J Infect Dis 2020; 98: 130–137. DOI: 10.1016/j.ijid.2020.06.046.

20.

Kamali

Nunn

Mulder

, et al. Seroprevalence and incidence of genital ulcer infections in a rural Ugandan population. Sex Transm Infect 1999; 75: 98–102. DOI: 10.1136/sti.75.2.98.

21.

Sudenga

Kempf

McGwin

Jr. , et al. Incidence, prevalence, and epidemiology of herpes simplex virus-2 in HIV-1-positive and HIV-1-negative adolescents. Sex Transm Dis 2012; 39: 300–305. DOI: 10.1097/OLQ.0b013e318244a90f.

22.

Phipps

Saracino

Magaret

, et al. Persistent genital herpes simplex virus-2 shedding years following the first clinical episode. J Infect Dis 2011; 203: 180–187. DOI: 10.1093/infdis/jiq035.

23.

World Health Organisation . HIV/AIDS, https://www.who.int/news-room/fact-sheets/detail/hiv-aids (2022, accessed 10 June 2022).

24.

Asare

Osman

Ngcapu

, et al. Burden of sexually transmitted infections from acute HIV infection among women in South Africa: evidence from a prospective cohort study. Ann Epidemiol 2022; 74: 132–139. DOI: 10.1016/j.annepidem.2022.08.038.

25.

Centers for Disease Control and Prevention . Sexual risk behaviors, https://www.cdc.gov/healthyyouth/sexualbehaviors/index.htm (2023, accessed 28 March 2023).

26.

Langen

. Gender power imbalance on women’s capacity to negotiate self-protection against HIV/AIDS in Botswana and South Africa. Afr Health Sci 2005; 5: 188–197. DOI: 10.5555/afhs.2005.5.3.188.

27.

Masson

Arnold

Little

, et al. Inflammatory cytokine biomarkers to identify women with asymptomatic sexually transmitted infections and bacterial vaginosis who are at high risk of HIV infection. Sex Transm Infect 2016; 92: 186–193. DOI: 10.1136/sextrans-2015-052072.

28.

Nagot

Ouedraogo

Defer

, et al. Association between bacterial vaginosis and herpes simplex virus type-2 infection: implications for HIV acquisition studies. Sex Transm Infect 2007; 83: 365–368. DOI: 10.1136/sti.2007.024794.

29.

Esber

Vicetti Miguel

Cherpes

, et al. Risk of bacterial vaginosis among women with herpes simplex virus type 2 infection: a systematic review and meta-analysis. J Infect Dis 2015; 212: 8–17. DOI: 10.1093/infdis/jiv017.

30.

Abbai

Nyirenda

Naidoo

, et al. Prevalent herpes simplex virus-2 increases the risk of incident bacterial vaginosis in women from South Africa. AIDS Behav 2018; 22: 2172–2180. DOI: 10.1007/s10461-017-1924-1.

31.

Christine

Amalia

Krista

, et al. P2.11 Association between genital herpes simpex virus type-2 shedding and presence of bacterial vaginosis-associated bacteria. Sex Transm Infect, 2017; 93(Suppl 2): A74.3–A75.

32.

Torcia

. Interplay among vaginal microbiome, immune response and sexually transmitted viral infections. Int J Mol Sci 2019; 20: 266. DOI: 10.3390/ijms20020266.

33.

Dorward

Sookrajh

Gate

, et al. HIV treatment outcomes among people with initiation CD4 counts >500 cells/µL after implementation of treat all in South African public clinics: a retrospective cohort study. J Int AIDS Soc 2020; 23: e25479. DOI: 10.1002/jia2.25479.

34.

Phillips

Clancy

. Some effects of “social desirability” in survey studies. Am J Sociol 1972; 77: 921–940.

35.

Centers for Disease Control and Prevention . Principles of epidemiology in public health practice. 3rd ed. Atlanta, USA: Centers for Disease Control and Prevention (CDC), 2012.

Incidence of herpes simplex virus type 2 positivity among women living with human immunodeficiency virus in South Africa

Abstract

Background

Methods

Results

Conclusion

Keywords

Introduction

Methods

Study design, population, and setting

Primary outcome

Exposure variables

Laboratory measures

Data collection

Statistical analyses

Results

Characteristics of participants at enrolment

Incidence of HSV-2 PCR positivity after acute HIV infection among women

Factors associated with time to HSV-2 PCR positivity after acute HIV infection among women

Cox failure curves showing the adjusted probability of HSV-2 PCR positivity over time

Multivariable Cox regression model goodness of fit: the proportional-hazards assumption

Discussion

Footnotes

Acknowledgements

Declaration of conflicting interests

Funding

Ethical statement

Contributors

ORCID iD

References