Leutscher, Peter MD, PhD*†; Jensen, Jorgen Skov MD, PhD‡; Hoffmann, Steen MD‡; Berthelsen, Lene CRT‡; Ramarakoto, Charles-Emile MD*; Ramaniraka, Vero MD*; Randrianasolo, Bodo MD*; Raharisolo, Clairette MD*; Böttiger, Blenda MD, DMSc‡; Rousset, Dominique MD, PhD*; Grosjean, Pierre MD*; McGrath, Moriah McSharry MPH§; Christensen, Niels DMSc†; Migliani, Rene MD, PhD*
THE HIV PANDEMIC had a delayed arrival in Madagascar due to the island’s geographical separation from the African continent. However, seroprevalence studies have shown that HIV has by now spread to all of the country’s 6 provinces.1–6 Although national surveillance data are unable to accurately gauge the HIV prevalence on this remote island, it is estimated at 1%, high enough to indicate an early-stage epidemic requiring immediate intervention.
Management of sexually transmitted infections (STIs) is an essential step in HIV control. STIs and HIV share many socioeconomic and epidemiologic dimensions, and STI is a known cause of increased risk for HIV transmission during sexual intercourse.7–10 Studies on STIs and HIV prevalence in Madagascar have so far focused on patients attending urban clinics,1–5,11–15 whereas the situation in rural Madagascar has not received the same attention. It is of paramount importance to collect data concerning STI prevalence and sexual behavior in rural areas, where more than 70% of the Malagasy population resides. The status of rural settings cannot be extrapolated from information on STIs and HIV because of radically different infrastructures and socioeconomic environments. Rural residents have extremely minimal access to the educational, medical, and communication resources that are available to many city dwellers. Therefore, the current study was undertaken to provide information regarding the STI and HIV situation in one rural area in Madagascar.
The study was conducted in 2002 among adults aged 15 to 49 years living in the rural villages of Madagascar’s northwest coast. Participants were recruited from the Mataipako village and from 4 representatives of 16 villages of the 10,000-ha Sirama sugarcane plantation: Ambodikatakata, Ambodimanga, Ankatoko, and Tanambao. Mataipako is home to a private general practitioner, and the residents have access to a government dispensary 20 km further south on the main road. The main source of income is cultivation of rice, manioc, beans, and cashew nuts. In Sirama, plantation workers live with their families in villages, where limited and very basic health care is provided to employees and their dependents at the plantation company dispensaries. Those who are not eligible to use the plantation clinics travel up to 15 km to public facilities outside the plantation limits. The population of the study villages varies from 500 to 1100. The small downtown area of Sirama (which is 1 to 12 km from different villages) hosts the factory facilities and company offices, as well as local commerce, schools, and the main health center, which includes a hospital and a prenatal clinic. Sugarcane products, such as raw sugar and alcohol distilled from molasses, are distributed overland by the road to Ambilobe or by sea from the harbor, Port Louis, in Ansohimbondrona, 12 km west of the Sirama downtown.
Participants were examined for STIs as part of a cross-sectional study on urogenital schistosomiasis. Information regarding sexual history and male condom use was obtained from each participant using a standardized questionnaire. Individuals who had not reached sexual debut were excluded from the study. The social and cultural aspects of the study population were observed and documented by a US Peace Corps volunteer, a member of the research team, who lived on site in Sirama for 2 years surrounding data collection.
Urogenital specimens were obtained using charcoal-impregnated cotton-tipped swabs on plastic shafts. Cervical swabs from women and urethral swabs from men for diagnosis of Neisseria gonorrhoeae (Ng) were stored in cryotubes (Nunc, Roskilde, Denmark), and within 1 hour of sampling placed in liquid nitrogen at −193°C until culture could be performed on selective plates. The E-test (AB-Biodisk, Solna, Sweden) was used for minimal inhibition concentration determination. Betalactamase production was tested with a chromogenic cephalosporin method.
First-void urine specimens from women and men were examined for Chlamydia trachomatis (Ct) and Mycoplasma genitalium (Mg). Ct was detected by an in-house polymerase chain reaction (PCR) assay detecting the cryptic Ct plasmid, and positive results were confirmed by a second PCR detecting the Ct 16S rRNA gene.16–18 Mg was detected using an in-house PCR by amplifying parts of the 16S rRNA gene, and positive results were confirmed by amplification of a fragment of the Mg-Pa adhesin gene.19,20 All primary PCR assays incorporated an internal control for inhibition.
Trichomonas vaginalis (Tv) was detected in the vaginal and male urethral swabs using the InPouchTV culture system (BioMed Diagnostics, Santa Clara, CA).21 Bacterial vaginosis (BV) was diagnosed by wet smear using the criteria of Amsell et al.22 Three criteria should be present for the diagnosis of yeast vaginitis (YV): (1) presence of pseudohyphae; (2) pH of vaginal fluid <5.0; and (3) watery or white discharge with cheesy plaques.
A blood sample was obtained from each study participant to test for Treponema pallidum (Tp) antibodies with a nontreponemal antigen test (RPR, Slide test; Biomérieux, Marcy l’Etoile, France) and an indirect hemagglutination test (TPHA-TB; Technique Biologique, Paris, France). Detection of antibodies against herpes simplex virus (HSV) types 1 and 2 was done by an indirect blocking ELISA as earlier described,23 where the type specificity was improved by blocking of HSV-type common epitopes with type-heterologous rabbit immunoglobulins.24 Sera were screened anonymously and masked to personal identifiers for HIV antibodies with an ELISA (Genscreen HIV 1/2, version 2; Bio-Rad, Marnes la Coquette, France) and Western blotting (New Lav Blot 1 and New Lav Blot 2; Bio-Rad).
Baseline Anti-STI Treatment and Follow-up
Anti-STI treatment was systematically given to all participants upon baseline examination. The national STI regimen recommended by Ministry of Health in Madagascar (1 oral dose of 500 mg ciprofloxacin (Ci), and 100 mg doxycycline (D), orally twice daily for 7 days) was compared with an alternative 1-day regimen. Sixty percent of the women were, according to the study protocol, randomized to receive the national CiD regimen and 40% the alternative regimen. The randomization for men was 50% for each regimen. Pregnant and breast-feeding women initially randomized for the CiD regimen were moved to the second randomized group (ie, the alternative 1-day CeA regimen group,) receiving 1 intramuscular dose of ceftriaxone (Ce) 250 mg and 1 oral dose of 1 g azithromycin (A). The alternative 1-day regimen for men consisted of 1 oral dose of 500 mg Ci and 1 oral dose of 1 g azithromycin. Metronidazole, 1 oral dose of 2 g, was added to both the national and the alternative regimens to cover Tv and BV. However, pregnant women in the first trimester (as confirmed by ultrasonography) did not receive metronidazole. YV was treated with clotrimazole 10% vaginal cream for 1 week. Individuals with positive results for both RPR and TPHA tests were treated with a single dose of benzathine penicillin (2.4 million international units). Partners who were not enrolled in the study were also offered STI treatment. Cure and reinfection rates were assessed 3 weeks and 6 months, respectively, after baseline treatment.
Ethics and Statistics
The study was approved by the National Committee on Ethics in Madagascar. Participation in the study required a written informed consent from each individual. Due to extremely low literacy, researchers obtained verbal consent from participants before asking for their signature. Participants who did not know how to write indicated consent with an “X” on the form, drawn under observation by the researcher. Participants were informed that they could at any point withdraw from the study. For data analysis purposes, participants were divided into 3 age groups: 15 to 24 years, 25 to 34 years, and 35 to 49 years. ÷2 And Fisher test were used when appropriate to compare proportions and associations in bivariate analyses as part of the Epi-Info (version 6.0) data handling system.
A complete set of data was obtained from 643 (89%) of 727 individuals initially enrolled in the study (333 [52%] women and 310 [48.2%] men). Forty-eight percent of the study participants were from the Mataipako village, and 52% were from the Sirama villages. Median age was 29 years for both women and men in the 2 study areas.
The overall Ng, Ct, Mg, Tv, and Tp prevalence, expressed as the percentage of the study population infected with one or more of these pathogens, was significantly higher in women (38%) than in men (27%) (P < 0.01; OR, 0.86; 95% CI, 0.43–0.86) (Table 1). In women, young age (15–24 years) as compared with older age (25 to 49 years) was associated with infection (P < 0.01; OR, 2.00; 95% CI, 1.23–3.27). In men, the overall prevalence was around 27% in all 3 age groups. Ng, Ct, and/or Mg were detected in 84 (13%) individuals (50 [15%] women and 34 [11%] men). The prevalence was 13% in both the Mataipako village and the Sirama villages. In bivariate analyses, Ng, Ct, and/or Mg infection was associated with young age (15–24 years) both in women (P < 10−6; OR, 11.02; 95% CI, 3.55–38.2) and in men (P < 0.01; OR, 4.94; 95% CI, 1.63–15.95) (Table 2).
Ng was isolated by culture in 18 (5%) of the women and 10 (3%) of the men. Methylene blue–stained smears from 243 randomly selected individuals (124 women and 119 men) with a negative culture were examined by an experienced laboratory technician for quality control purposes. Diplococci resembling gonococci were detected in 27% and 29% of the smears from the women and the men, respectively.
The Ct prevalence was significantly higher in women (9.3%) than in men (4.5%) (P < 0.02; OR, 0.49; 95% CI, 0.26–0.89) (Table 1). In contrast to Ng and Ct, Mg was more prevalent in men (5%) than in women (3%), although the difference did not reach significance. Seven (14%) women and 5 (15%) men infected with Ng were also infected with Ct and/or Mg.
Tv was significantly less prevalent in men (7.4%) than in women (23%) (P < 10−6; OR, 0.26; 95% CI, 0.15–0.44). The prevalence decreased by age in women, whereas the tendency was opposite in men, with a prevalence of 3% in the 15- to 24-year group compared with 13% in the 35- to 49-year group (P < 0.05; OR, 0.25; 95% CI, 0.05–0.96) (Table 1). BV was diagnosed in 110 (32%) women. The overall Tv and BV prevalence was 38% (87 of 231) in women aged 15 to 34 years in comparison with 23% (23 of 102) in women aged 35 to 49 years (P < 0.01; OR, 2.08; 95% CI, 1.18–3.68). YV was diagnosed in 20 (6%) women, and without any age-depending difference.
Tp was diagnosed in 33 (11%) men and 29 (9%) women. In both sexes, Tp was more prevalent in individuals aged 25 to 49 years (12% in women and 9% in men) than in the individuals aged 15 to 24 years (4% in women and 5% in men) (P < 0.05). HSV type 1 antibodies were detected in 317 (95%) women and in 305 (98%) men. There was no significant difference in the prevalence between the different age groups. The HSV type 2 antibody prevalence was significantly higher in women than in men: 49% (n = 162) and 28% (n = 88), respectively (P < 10−5; OR, 2.39; 95% CI, 1.70–3.36). In women, 32 (28%) individuals in the 15- to 24-year age group tested positive for HSV type 2 antibodies compared with 64 (56%) individuals in the 25- to 34-year age group (P < 0.0001; OR, 0.30; 95% CI, 0.16–0.57). In the 35- to 49-year age group, 66 (65%) women tested positive. In men, 13 (14%) individuals in the youngest age group were HSV-2 antibody positive compared with the 25- to 34-year age group, where 31 (30%) were positive (P < 0.001; OR, 0.35; 95% CI, 0.15–0.87). In the 35- to 49-year age group, 44 (39%) men tested positive. Six (1%) individuals tested positive for HIV antibodies.
A history of multiple partners, defined as 2 or more different partners in the last 3 months, was reported by 28 (8%) women and by 141 men (46%) (P < 10−6; OR, 0.18; 95% CI, 0.07–0.18). Multiple partners were reported by 15 (13%) women aged 15 to 24 years and by 4 (4%) women in the age group 35 to 49 years (P < 0.02; OR, 3.30; CI, 1.13–9.62), whereas it was reported by 52 (54%) men aged 15 to 24 years and by 43 (38%) of those aged 35 to 49 years (P < 0.03; OR, 1.90; CI, 1.05–3.43). A history of multiple partners was associated with Ng, Ct, and/or Mg infection in women (P < 0.001; OR, 0.23; 95% CI, 0.09–0.56) but not in men (P = 0.19) (Table 2).
Women reported significantly less frequent condom use in the last 3 months than did men: 19 (6%) women versus 50 (16%) men (P < 0.0001; OR, 3.18; 95% CI, 1.77–5.71). Condom use was less commonly reported by women aged 15 to 24 years (3%) than by women aged 25 to 49 years (7.0%). The frequency of male condom use was equal in the 3 age groups. Condom use at last sexual intercourse with an incidental partner was reported by 12 (4%) of the women and by 22 (7%) of the men. Reported use of condoms in the last 3 months was not associated with a significantly lower Ng, Ct, and/or Mg infection prevalence (Table 2).
Complete posttreatment data were obtained from 265 (41%) individuals (133 women and 132 men) at the 3-week follow-up and from 461 (72%) at the 6-month follow-up (Table 3). The Ng, Ct, and/or Mg prevalence declined in the women from 15% at baseline to 2% 3 weeks later, and in men from 11% to 1%. Two women diagnosed with Mg and both treated with the CiD regimen at baseline were rediagnosed with Mg at the 3-week follow-up. The 6-month follow-up survey revealed an increase in overall prevalence but most prominently in the women. Ng, Ct, and/or Mg were detected in 20 (8%) women and in 5 (2%) men (P < 0.01; OR, 0.09; 95% CI, 0.09–0.79). Six (30%) of the 20 women and 1 (20%) of the 5 men had also been diagnosed with Ng, Ct, and/or Mg infection at baseline. The 6-month posttreatment prevalence was 19% (14 of 74) in women aged 15 to 24 years versus 3% (6 of 175) in the 25- to 49-year age group (P < 0.0001; OR, 6.57; 95% CI, 2.23–20.20). Tv was diagnosed in 31 (13%) women and in 6 (3%) men. Tv prevalence was 20% in women aged 15 to 24 years compared with 9% in women aged 25 to 49 years (P < 0.02; OR, 2.53; 95% CI, 1.10–5.79). BV and YV were diagnosed at follow-up in the same proportions as at baseline: BV (31%) and YV (10%).
The STI prevalence among this rural study population in the northwestern coastal region of Madagascar was as high as previously observed among selected patient groups attending urban STI clinics.11,13,15 The findings strongly suggest that implementation of STI control programs is urgently warranted in this area and other coastal rural areas of the island.
The baseline prevalence and the 6-month posttreatment infection rate of Ng, Ct and/or Mg was highest in individuals aged 15 to 24 years, particularly among women, supporting the current consensus that young adults should constitute a priority target group in STI control programs. These control programs, as well as behavioral interventions, must take into account the cultural dimensions of STI epidemiology. In Sirama, it is common practice for older men to partner with younger women, who are less likely to have the social and economic resources to prevent and treat STIs. Madagascar is similar to many other countries in the third world where men have more access to the household economy, although the women are very strongly present in Sirama’s family structure and even in its paid workforce. Women are often reliant on their male partners for household income, and this is perpetuated by the custom in which women receive gifts after having intercourse with a man. These and other aspects of the coastal Malagasy sexual economy, which is distinct from what occurs in the central highlands and capitol region, exacerbate the disease burden on communities like those studied here.
It could be argued that the real Ng, Ct and/or Mg prevalence may have been higher than detected. First, diplococci by microscopy were observed at a higher frequency by microscopy than was found in culture, which suggests that Ng may have died in a proportion of specimens during the storage process between sampling and cultivation. Second, examining only urine samples for detection of Ct and Mg in women may lead to an underestimate of the prevalence since the sensitivity of this specimen type is approximately 90% for both pathogens as compared with examination of cervical swab specimens.18 It was furthermore shown in the same study that urine frozen before analysis may also lead to a decreased detection rate. Third, a high proportion of women were observed with BV, which in other studies has been associated with a lower prevalence of Mg.25 This observation may explain the rather low prevalence of Mg in women in this study compared to that of Ct, which has been detected more often in women with BV.26
Tv was diagnosed in one fourth and BV in one third of the women, and both infections were more prevalent in younger women. The implications of Tv and BV are potentially much more severe than simple vaginal discharge. They are also considered risk factors for HIV transmission, as well as potential causes of complications during pregnancy.27–29
Tp had a different age distribution in both women and men than did Ng, Ct and/or Mg, since it was more prevalent in individuals aged 25 to 49 years than in those aged 15 to 24 years. The observed Tp prevalence is based on positive results of both the VRDL test and the TPHA test. Yaws is, according to local health authorities, not encountered in this part of Madagascar. Reliable serology screening and treatment protocols, including partner treatment, during antenatal care are essential tools for reducing the imminent risk of adverse pregnancy outcome associated with Tp.30–32
The prevalence of HSV-2 infection in our study of a population in a rural area of Madagascar is very similar to that found in a study of rural areas in Tanzania, where 44% of the women and 24% of the men had antibodies to HSV-2.33 The concordant findings suggest that women are more receptive to infection by the HSV-type virus than men. We also found a significant association between HSV-2 seroprevalence and age, which also is described as a universal finding in the review of HSV-2 epidemiology.34 Infection with HSV-2 is generally an STI and a major cause of genital ulcers. The damage of the mucosal membrane provides a direct entry for HIV, and genital ulcers are a known risk factor for HIV transmission in Africa. The high prevalence of HSV-2 infection among the rural population in Madagascar is an important finding, stressing the need for implementing preventive measures for STIs, including HIV infection, among this population.
This study population was, in addition to showing high STI prevalence in younger individuals, characterized by critical risk factors such as multiple partners and infrequent use of condoms. Clearly, lack of education in addition to inadequacies in infrastructure, including health facilities, contribute to the high STI endemicity in the study area. A health interview survey conducted during the study revealed that men and women in Sirama are for most part unaware of the symptomatology and transmission of STIs.
When STI is discussed with women, it is seen as a marker of an inevitable cultural phenomenon: that men are maditry (mischevious, naughty). According to the women of Sirama, men are unable to confine their sexual activity to 1 partner. Even when a man is faithful to his wife (and most men are married), he is likely to soon have another partner, as marriages in the region are rarely legalized and often short-lived. Though most men and women in the study villages were familiar with social marketing campaigns of NGO condom distributors, most preferred dancing to pop songs that mention the prophylactics to using them. There is in general a well-entrenched antipathy to use condom. First, many women and men still wish to have children. Less obvious to the outsider is the Malagast notion that intercourse with a condom is tsy matsiro (not flavorful, devoid of pleasure). There has been little incentive to change this behavior because men and women do not realize how STI affects their health in the short and long term, and the late-arriving and poorly monitored HIV epidemic has yet to resonate with the Malagasy people. Furthermore, there is no visible presence of satisfied customers: men and women who speak out about the benefits they experience in using condoms.
There are also local approaches to treating STIs, which may dampen the appeal of Western biomedical prevention and treatment. STIs in Sirama may be treated in the home with herbal infusions, known as tisane, aody (medicine), and makefy (bitter). These techniques are usually a complement to conventional medical treatment but are heavily used in cases of STIs, as infected individuals may fear stigmatization with more public treatment. Incorrect treatment with antibiotics also contributes to wide STI distribution. Procured in local pharmacies or general stores, pills are often mislabeled and/or stored improperly, with “prescription” regimens determined by nonclinicians. The research team witnessed many antibiotic sales by local providers and took many histories of patients who had followed haphazard courses of antibiotics.
Unfortunately, even treatment at health care facilities is imperfect. Though STI prescriptions in general are handled by physicians employed by the plantation company, it was observed during the study that none of these clinicians made use of the syndrome management treatment algorithm adopted and recommended by the Ministry of Health. Because partner notification is spotty and hardly ever takes into account more than 1 partner, the patient is consistently reexposed. This engenders distrust of the dispensary, since medication appears to be ineffective when follow-up brings a repeat diagnosis. Long waits and high costs are primary deterrents to use of the company clinic, but the way the patients are treated disinclines them to present for screening, treatment, or education.
Though STIs are extremely prevalent and morbidity, including the shameful burden of infertility, is high, educational and public health systems have made little headway in empowering the rural population to combat STIs. The reproductive health of the Malagasy population will continue to be compromised by STIs, until the Malagasy government, with appropriate technical assistance from donor organizations, has adopted a new public health paradigm in which existing control measures are expanded beyond the urban areas and selected high-risk groups of prostitutes to also cover the rural coastal areas. However, the central political and administrative innovations must also be accompanied by intensive capacity building of community leaders at the local level.
Nearly 1% of the study population was tested positive for HIV antibodies. Our observations indicate that the Malagasy population is extremely vulnerable to being devastated by an HIV epidemic within the next decade. In addition to adjustment of care-seeking and sexual behavior, appropriate intervention must promote early diagnosis and treatment of STIs. This is crucial in early-stage HIV epidemics,35–38 as is the case in Madagascar.
1. Harms G, Kirsch T, Rahelimiarana H, et al. HIV and syphilis in Madagascar. AIDS. 1994;8:279–280.
2. Zeller H, Ramamonjisoa A, Bosier P, et al. HIV infection in Madagascar in 1995. AIDS. 1997;11:401–402.
3. Behets F, Andriamahenina R, Andriamiadana J, et al. High syphilis and low but rising HIV seroprevalence rates in Madagascar. Lancet. 1996;347:831.
4. Andriamahenina R, Ravelojaona B, Rarivoharilala E, et al. Le SIDA à Madagascar: épidémiologie, projections, impact socio-économiques, interventions. Bull Soc Pathol Exot. 1998;91:68–70.
5. Champetier de Ribes G. MST/SIDA en 1998: données du dispensaire 67 ha d’Antananarivo, soutenu par Médecins du monde. Bull Info Epi Santé Publique. 1999;15:4–5.
6. Leutscher P, Behets F, Rousset D, et al. Sexually transmitted infections and sexual practice among males in different rural settings in Madagascar: implications of HIV propagation on the island. Sex Transm Dis. 2003;30:262–265.
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. Weir SW, Feldblum PJ, Roddy RE, et al. Gonorrhea, as a risk factor for HIV acquisition. AIDS. 1994;8:1605–1608.
9. Plourde PJ, Pepin J, Agoki E, et al. Human immunodeficiency virus type 1 seroconversion in women with genital ulcers. J Infect Dis 1994;170:313–7.
10. Cohen MS, Hoffmann IF, Royce RA, et al. Reduction of concentration of HIV-1 in semen after treatment of urethritis: implications for prevention of sexual transmission of HIV-1. Lancet. 1997;349:1868–1873.
11. Harms G, Matull R, Randrianasolo D, et al. Pattern of sexually transmitted diseases in a Malgasy population. Sex Transm Dis. 1994;21:315–320.
12. Gateau T, Zeller HG. Approche épidémiologique des maladies sexuellement transmissibles (MST) á Antsiranana (Nord de Madagascar): entre préventif et curatif, le choix d’une stratégie face aux MST. Arch Inst Pasteur Madagascar. 1996;63:8–11.
13. Behets FMT, Andriamiadana J, Randrianasolo D, et al. Chancroid, primary syphilis, genital herpes, and lymphogranuloma venereum in Antananarivo, Madagascar. J Infect Dis. 1999;180:1382–1385.
14. Gleize L, Randramanga R, Ratsimbazafy N, et al. Prise en charge des maladies sexuellement transmissibles par l’approche et le dépistage VIH volontaire dans un dispensaire spécialisé d’Antananarivo (Madagascar). Arch Inst Pasteur Madagascar. 2000;66:46–49.
15. Behets FMT, Andriamiadana J, Randrianasolo D, et al. Laboratory diagnosis of sexually transmitted infections in women with genital discharge in Madagascar: implications for primary care. Int J STD AIDS. 2002;13:606–611.
16. Loeffelholz MJ, Lewinski CA, Silver SR, et al. Detection of Chlamydia trachomatis in endocervical specimens by polymerase chain reaction. J Clin Microbiol. 1992;30:2847–2851.
17. Pollard DR, Tyler SD, Ng CW, et al. A polymerase chain reaction (PCR) protocol for the specific detection of Chlamydia spp. Mol Cell Probes. 1989;3:393–399.
18. Jensen JS, Björnelius E, Dohn B, et al. Comparison of first void urine and urogenital swab specimens for detection of Mycoplasma genitalium and Chlamydia trachomatis by polymerase chain reaction in patients attending a sexually transmitted disease clinic. Sex Transm Dis. 2004;31:499–507.
19. Jensen JS, Borre MB, Dohn B. Detection of Mycoplasma genitalium by PCR amplification of the 16S rRNA gene. J Clin Microbiol. 2003;41:261–266.
20. Jensen JS. Polymerase chain reaction for detection of Mycoplasma genitalium in clinical samples. J Clin Microbiol. 1991;29:46–50.
21. Borgchardt KA, Smith RF. An evaluation of an InPouch TV culture method for diagnosing Trichomonas vaginalis infection. Genitourin Med. 1991;67:149–152.
22. Amsel R, Totten PA, Spiegel CA, et al. Non-specific vaginitis: diagnostic criteria and microbial and epidemiologic associations. Am J Med. 1983;74:14–22.
23. Vestergaard BF, Grauballe PC, Spanggaard H. Titration of herpes simplex virus antibodies in human sera by the enzyme-linked immunosorbent assay (ELISA). Acta Pathol Microbiol Immunol Scand. 1977;85:466–468.
24. Vestergaard BF, Grauballe PC. ELISA for herpes simplex virus (HSV) type-specific antibodies in human sera using type-heterologous rabbit antibodies. Acta Pathol Microbiol Immunol Scand. 1979;87:261–263.
25. Keane FE, Thomas BJ, Gilroy CB, et al. The association of Mycoplasma hominis, Ureaplasma urealyticum and Mycoplasma genitalium with bacterial vaginosis: observations on heterosexual women and their male partners. Int J STI AIDS. 2000;6:356–360.
26. Wiesenfeld HC, Hillier SL, Krohn MA, et al. Bacterial vaginosis is a strong predictor of Neisseria gonorrhoeae and Chlamydia trachomatis infection. Clin Infect Dis. 2003;36:663–668.
27. Cotch MF. Vaginal infection and prematury study group: carriage of Trichomonas vaginalis is associated with adverse pregnancy outcome. The 36th Interscience Conference on Antimicrobial Agents and Chemotherapy, Washington, DC: American Society for Microbiology, 1990.
28. Cohen CR, Duerr A, Pruithithada N, et al. Bacterial vaginosis and HIV seroprevalence among female commercial sex workers in Chiang Mai, Thailand. AIDS. 1995;9:1093–1097.
29. Sewankambo N, Gray RH, Wawer MJ, et al. HIV-infection associated with abnormal vaginal flora morphology and bacterial vaginosis. Lancet. 1997;350:546–550.
30. Donders GG, Desmyter J, De Wet DH, et al. The association of gonorrhoea and syphilis with premature birth and low birthweight. Genitourin Med. 1993;69:98–101.
31. McDermott J, Steketee R, Larsen S, et al. Syphilis-associated perinatal and infant mortality in rural Malawi. Bull World Health Organ. 1993;71:773–780.
32. Watson-Jones D, Gumodoka B, Weiss H, et al. Syphilis in pregnancy in Tanzania, II: the effectiveness of antenatal syphilis and single-dose benzathine penicillin treatment for the prevention of adverse pregnancy outcomes. J Infect Dis. 2002;168:948–957.
33. del Mar Pujades Rodriguez M, Obasi A, et al. Herpes simplex virus type 2 infection increases HIV incidence: a prospective study in rural Tanzania. AIDS. 2002;16:451–462.
34. Weiss H. Epidemiology of herpes simplex virus type 2 infection in the developing world. Herpes. 2004;11(suppl 1):24–35.
35. Grosskurth H, Mosha F, Todd J, et al. Impact of improved treatment of sexually transmitted diseases on HIV-1 infection in rural Tanzania: a randomized controlled trial. Lancet. 1995;346:530–536.
36. Wawer MJ, Sewankambo N, Serwadda D, et al. Control of sexually transmitted diseases for AIDS prevention in Uganda: a randomised community trial. Lancet. 1999;353:525–535.
37. Hudson CP. Community-based trials of sexually transmitted disease treatment: repercussions for epidemiology and HIV treatment. Bull World Health Organ. 2001;79:48–59.
38. Kamali A, Quigley M, Nakiyingi J, et al. Syndromic management of sexually transmitted infection and behaviour change interventions on transmission of HIV-1 in rural Uganda: a community randomized trial. Lancet. 2003;361:645–652.