HIV TRANSMISSION THROUGH HETEROSEXUAL INTERCOURSE was recorded in Chennai, Tamilnadu, India, in the late 1980s,1 followed by reports of HIV among men having sex with men2 and injecting drug users (IDUs).3 In response, a number of targeted interventions for “groups at risk” were launched including harm-reduction initiatives for IDUs. Sentinel surveillance in Chennai since 1999, however, consistently recorded HIV prevalence over 20% among IDUs with a current prevalence rate of 33% in 2005.4 Behavioral data from IDUs in Chennai5 indicated that epidemiological understanding of the sexually transmitted infections (STIs) among them and their regular sex partners would be necessary. Given the known transmission synergy between HIV and other STIs,6–7 this understanding is important for HIV intervention development. We therefore carried out a cross-sectional study in Chennai.
Materials and Methods
Peer outreach workers and field researchers contacted IDUs and their regular sex partners over a period of 3 months from April to July 2003. All the IDUs were males and none of the female participants had ever injected drugs. After informed consent, participants were interviewed with a semistructured questionnaire on drug use history, sexual practices, and physical illnesses. The study received ethical approval of the Institutional Review Boards of the Population Council, NY, and the nongovernmental organization implementing the study.
Each participant provided a specimen of blood and serum was examined for herpes simplex type 2 infections using an HSV-2 ELISA test system (Focus Technology, USA) and for HIV antibodies using two different ELISA tests (Biomeuriex and Vironostika, USA). In addition, each IDU provided a sample of first-void urine after abstaining from urination for 2 hours for NG/CT (Neisseria gonorrhea and Chlamydia trachomatis) duplex PCR tests (Roche Molecular system Inc., Germany). Two vaginal swabs collected from each of the consenting female participants were examined—one used for identifying motile trophozoites of Trichomonas vaginalis (TV) in wet mount and yeast cells and the other for diagnosing bacterial vaginosis (BV) by using Nugent's criteria on gram stained smears. Cervical swab collected from each of the female participants was examined by NG/CT duplex PCR tests.
Participants received support in accessing HIV voluntary counseling and testing and treatment for clinical genital syndromes and laboratory-confirmed infections free of cost. We used SPSS (version 13.0), Epi-Info (version 6.4b) and I-stat for data analysis.
Of the 260 male IDUs and their regular female sex partners recruited for the study, behavioral and HIV prevalence data were available for 226 couples. In 11 of the 226 couples (5%), both IDUs and their regular female sex partners were HIV infected.8 As no other woman was HIV-positive, it was not possible to examine the determinants for HIV transmission with the small sample of 11 HIV-positive women. Given that 30% of the male IDUs (68/226) were HIV infected, factors responsible for occurrence of STIs in women were, however, investigated to understand the likelihood of acquiring HIV infection in future in the HIV-negative women (N = 215). Data on STIs was incomplete in four HIV-negative women, and hence, the analysis was done on a complete set of behavioral and biologic data available for 211 women and their male IDU partners. While the mean age of IDUs was 36 years (range 22–52 years; SD ±6; median 36), the mean age of their regular sex partners was comparatively lower (30 years; range 18–45 years; SD ±6; median 30).
Sexually Transmitted Infections
None of the IDUs and their regular sex partners was infected with NG. CT was detected in only one female. Test for syphilis was positive (indicated by both RPR and TPHA test) in only 1% of the IDUs and 2% of their regular sex partners. Serodiscordance was observed in three couples where only IDUs were positive and in four couples where only female regular sex partners of IDUs were positive for syphilis. TV in wet mount was detected in 5% of the women (10/211) and 37% (77/211) of the women had laboratory-confirmed BV.
Prevalence of viral STIs was high in the study population; 40% of the IDUs (84/211) and 38% of their regular female sex partners (81/211) were infected with HSV-2. Serodiscordance was observed in HSV-2 test results; in 17% (36/211) couples only IDUs and in 16% (33/211) only regular female sex partners of IDUs were infected with HSV-2.
Self-Reported Sexual Practices and Condom Use
The mean and median age at sexual debut among IDUs was 19 years (range 10–29 years; SD ±3). Thirty-five percent of the IDUs reportedly had their first sex with female sex workers, and 41% of the IDUs (86/211) had it with casual partners or friends; the rest (24%) had first sex with their wives. The mean and median age at first sex among regular female sex partners of IDUs was lower at 18 years (range 12–33 years; SD ±4) compared to IDUs; almost all of them (204/211) reported first sex with husbands, five with friends, and two with casual partners. Eighty-two percent of the IDUs (172/211) and 6% (12/211) of the women reported sex with someone apart from their spouses.
A total of 113 IDUs (113/211; 51%) reported ever having sex with casual sex partners, of which 26% had it within the past 1 year. Of the IDUs, 62% (130/211) reported ever having sex with female sex workers; 16% (21/130) of them had sex in the past year. “Never using condoms” with sex workers, regular sex partners, and casual sex partners was reported by 62% (130/211), 87% (183/211), and 96% (108/113) of IDUs, respectively.
“Ever having sex with men” was reported by 22% of the IDUs (46/211). While 83% of this subpopulation had sex with men even before getting into injecting drug use, 51% had reported so after they took to injecting. Only one IDU reported “always” using condoms during same-gender sex. Two IDUs reported urethral discharge, and one IDU reported genital ulcer.
Univariate and Multivariate Analysis Findings
All the 211 HIV-negative regular female sex partners of IDUs were divided into two groups based on the presence of any STI (HSV-2, Chlamydia, Syphilis, TV, BV) in them. While 63% (133/211) of them had any STI, the rest (78/211; 37%) had none. Women having any STI were compared with those not having any STI on variables reflecting sociodemographic profile of women, their self-reported sexual practices, use of condom with the partners in regular relationship, self-reported sexual practices of IDUs, and presence of HIV or STIs in male IDU partners of women. The distributions, univariate OR, and CI of these variables are shown in Table 1. None other than “woman aged 38 and above,” “lower age at first sex of women,” “presence of HIV in male IDU partners” and “presence of coinfection of HIV and HSV-2 infection in IDUs” were significantly different between the two groups. Factors with biologic plausibility and having significance (P ≤0.1) in univariate analysis were subsequently entered in a multivariate logistic regression model to identify the variables associated with any STI in regular female sex partners of IDUs (Table 2).
In a multivariate model, women who had their first sex at age 17 years or less and those women who had HIV-positive IDUs as their regular male sexual partner had 2 times the odds of having any STI (OR 2.04; 95% CI 1.1–3.8; P = 0.02 and OR 2.38; 95% CI 1.1–5.1; P = 0.02 respectively). Women aged 38 and above in this model had seven times the odds of having any STI.
Through mapping and snowballing, we could only recruit 260 couples in the study, which was much lower than the expected number of 800, based on the impression of the field staff. Limiting the recruitment to central Chennai further meant that the sample was a convenient one. Though the IDUs in our study are similar to IDUs in other metropolitan cities of India in that their drug of choice for injecting is buprenorphine, they are predominantly males and are from lower socioeconomic backgrounds, different drug use pattern and dynamics in the northeastern states bordering Myanmar limits the wider applicability of the study findings. Despite this limitation and lack of data on sensitive issues such as concurrent sexual relationship, we believe the current study has generated useful information for intervention development.
First, HIV interventions among IDUs in India have largely focused on HIV transmission through sharing of syringes and needles. Only a few studies have looked at the risk of acquiring STIs by IDUs9 and their regular sex partners. However, there is some evidence that the sexual transmission of HIV in and from IDUs may be facilitated by presence of other sexually transmitted diseases. For example, in the northeastern state of Manipur, 45% of the noninjecting spouses of HIV-positive IDUs were found to be HIV infected, prevalence reached within 7 years of the onset of HIV epidemic among IDUs. The odds of HIV infection in wives of HIV-positive IDUs in this study was three times higher in presence of a history of a current sexually transmitted disease in either member of the married couples and two times higher when the duration of HIV infection in IDUs was greater than 8 years.10 In view of the synergy between other STIs and HIV,6–7 one would expect that rapid diagnosis and treatment of STIs would serve as cost-effective measures for slowing down the spread of HIV in this setting as seen elsewhere.11 Given the low rates of curable bacterial STIs and high HSV-2 prevalence in our study population, this approach, however, might require modification. Compared to the earlier work in Manipur, where around 10% of the IDUs had history of either urethral discharge or genital ulcers, only two males in the current study reported urethral discharge and one IDU genital ulcer. Moreover, older women (aged 38 and above) in our study were at risk of acquiring STIs, indicating the importance of reaching out to women in longer relationships with IDUs.
Serodiscordant results for syphilis as well as HSV-2 infection in couples indicate the potential for spread of STIs in and beyond the steady relationship. Stigma and potential consequences of reporting extramarital sexual activity may, however, have led to socially desirable responses by the women that could explain the mismatch between the self-reported behavioral data and infection prevalence. Ever use of condoms and consistent condom use with sex workers, regular sex partners, and casual partners was inadequate as in other studies in Asia12 and elsewhere.13 Moreover, 22% of the male IDUs reported ever having sex with men, and only one reported always using condoms during these sex acts.
Third, a high prevalence of BV (37%) in women in our study could also pave the way for HIV acquisition as highlighted by a study in women that identified an association between increased frequency of HIV infection and abnormal vaginal flora.14
We therefore conclude that the ongoing harm-reduction measures for IDUs based on needle syringe exchange, male condom promotion, and behavior change communication in India and other similar settings should also include their regular sex partners, have additional elements of outreach to women, and promotion of women-controlled safer sex measures. Managing STIs as indicated by baseline studies in IDUs and their regular sex partners should also form an integral part of this enhanced comprehensive intervention approach.
1. John TJ, Babu PG, Jayakumari H, et al. Prevalence of HIV infection in risk groups in Tamilnadu, India. Lancet 1987; 1:160–161.
2. Ravinathan R, Amudhamozhi R, Chitra PC, et al. HIV sero-prevalence among the homosexuals of Tamilnadu and effects of condom skill training on AIDS prevention [abstract]. Seventh International Conference on AIDS, Florence, Italy, June 16–21, 1991:314.
3. Joseph R. HIV Risk behavior and risk reduction among injecting drug users in treatment [dissertation]. Chennai: Dr MGR Medical University, 1996.
4. The Tamil Nadu HIV Sentinel Surveillance Report, Tamilnadu State AIDS Control Society (TNSACS), Government of Tamilnadu, India; 2005:48.
5. Kumar MS, Mudaliar S, Thyagarajan SP, et al. Rapid assessment and response to injecting drug use in Madras, South India. Int J Drug Policy 2000; 11:83–98.
6. Plummer FA, Simonsen JN, Cameron DW, et al. Cofactors in male–female sexual transmission of human immunodeficiency virus type 1. J Infect Dis 1991; 163:233–239.
7. Laga M, Manoka A, Kivuvu M, et al. Non-ulcerative sexually transmitted diseases as risk factors for HIV-1 transmission in women: results from a cohort study. AIDS 1993; 7:95–102.
8. Panda S, Kumar MS, Lokabiraman S, et al. Risk factors for HIV infection in injection drug users and evidence for onward transmission of HIV to their sexual partners in Chennai, India. J Acquir Immune Defic Syndr 2005; 39:9–15.
9. Bradshaw CS, Pierce LI, Tabrizi SN, et al. Screening injecting drug users for sexually transmitted infections and blood borne viruses using street outreach and self collected sampling. Sex Transm Infect 2005; 81:53–58.
10. Panda S, Chatterjee A, Bhattacharya SK, et al. Transmission of HIV from injecting drug users to their wives in India. Int J STD AIDS 2000; 11:468–473.
11. Gilson L, Mkanje R, Grosskurth H, et al. Cost-effectiveness of improved treatment services for sexually transmitted diseases in preventing HIV-1 infection in Mwanza region, Tanzania. Lancet 1997; 350:1805–1809.
12. Lau JT, Feng T, Lin X, et al. Needle sharing and sex-related risk behaviours among drug users in Shenzhen, a city in Guangdong, southern China. AIDS Care 2005; 17:166–181.
13. Plitt SS, Garfein RS, Gaydos CA, et al. Prevalence and correlates of Chlamydia trachomatis
, Neisseria gonorrhoeae
, Trichomonas vaginalis
infections, and bacterial vaginosis among a cohort of young injection drug users in Baltimore, Maryland. Sex Transm Dis 2005; 32:446–453.
14. Sewankambo N, Gray RH, Wawer MJ, et al. HIV-1 infection associated with abnormal vaginal flora morphology and bacterial vaginosis. Lancet 1997; 350:546–550.