Abstract
Dear Editor:
In Latin America, 1.4 million persons are infected with HIV-1 and approximately half live in Brazil. 1 The Brazilian AIDS epidemic is relatively stable, with an estimated adult prevalence of 0.6%. 2 Therefore, vulnerable groups such as injection drug users, men who have sex with men, and female sex workers (FSW) are core groups that affect the epidemiology of HIV-1. 1
FSW are considered a bridge for HIV-1 transmission from high-risk populations to the general population. Unsafe sexual behaviors, sexually transmitted infections (STI) other than HIV-1, and illicit drug use are common risk behaviors among FSW. 3,4 In Brazil, this group represents 1% of the Brazilian female population from 15–49 years of age. 3
HIV-1 shows a great genetic variability associated with transmission and disease progression, and strain heterogeneity may affect responses to antiretroviral therapies and impedes vaccine development efforts. 5 In Brazil, HIV-1 subtypes B, F1, and C cocirculate. However the distribution of these subtypes shows regional differences in Brazil, which has continental territorial dimension. 6,7
The Brazilian HIV-1 epidemic has been expanding towards the interior of the country and among young women. In 1983, the male:female HIV-1 infection ratio was 40:1 and this rate declined to 1.6:1 in 2009. 2 The first case of AIDS reported in Goiás occurred in 1984 and through June 2010, this state accounts for 35% (11,918/34,057) of AIDS cases in the central–western region. In the last 5 years, Goiás State has reported an average of 600 new cases of AIDS per year. 2 A previous investigation showed an HIV-1 prevalence of 0.09% among pregnant women who attended prenatal clinics in Goiânia, the capital of Goiás State. 8 However, there are no data on HIV-1 prevalence among vulnerable populations such as FSW. Thus, the aim of this study was to estimate the HIV-1 prevalence, identify risk behaviors, and to detect HIV-1 subtypes circulating among FSW recruited in Goiânia City.
Goiás State (population 5,849,105) the most populous state in central-western Brazil is located far from the HIV-1 epidemic epicenter (southeastern region). We carried out a cross-sectional study among FSW in Goiânia (1,256,514 inhabitants) between May 2009 and June 2010. Female sex workers were defined as women who exchanged sexual services for money. The eligibility criteria were women who exchanged sex for money within the last 30 days. Exclusion criteria were being younger than 18 years of age and self-reporting as transgender.
Data were collected using a respondent-driven sampling (RDS) method used to explore hidden populations. RDS is a chain-referral method in which participants recruit a limited number of additional participants. 3,7 The selection process began by identifying 7 FSW referred to as “seeds.” After completing the interview, each “seed” received three coupons. These coupons were given to other FSW known to the recruiter who were not their partners, relatives, or close friends. The new recruits were interviewed and in turn they recruited 3 additional participants. This process continued until 395 participants were enrolled. As an incentive, each participant received 6 public transportation tickets and 2 additional transportation tickets for each successfully recruited referral.
Informed consent was obtained prior to enrolling the FSW into the study. The study was approved by the “Comitê de Ética em Pesquisa do Hospital das Clínicas da Universidade Federal de Goiás” (protocol no. 001/09).
FSW were interviewed in person by two members of the research team, previously trained in order to guarantee consistency in asking questions. All participants answered a standardized questionnaire adapted from the FSW module of the Health International Behavioral Surveillance Surveys designed to capture data on demographic characteristics and risk behaviors.
After the interview, 10 mL of peripheral blood was collected from each FSW in order to test for the presence of anti-HIV type 1 and 2 antibodies by enzyme-linked immunossorbent assay (ELISA; Wiener Lab, Rosario, Argentina). Positive and borderline results were confirmed by Western blot (New Lav Blot I, BioRad, Marnes-la-Coquette, France).
For pol gene sequencing, plasma RNA was extracted, retrotranscribed into complementary DNA (cDNA), and the entire HIV-1 protease (PR) and reverse transcriptase (RT) fragment (approximately 750 bp) were amplified by nested polymerase chain reaction followed by direct sequencing, as described by Cardoso et al. 9 HIV-1 subtypes were identified using REGA tool version 2.0. and by phylogenetic analysis using the Neighbor-Joining (NJ) method under Kimura's 2-parameter correction method using MEGA4 software. Simplot software was used for HIV-1 intersubtype recombination analysis. The presence of transmitted drug resistance mutations was evaluated by the Calibrated Population Resistance tool and by International AIDS Society-USA (IAS-USA) major mutation lists. GenBank accession numbers of the sequences presented in this study are JX134040, JX134041, JX134042, JX134043.
Data were entered into the SPSS program (SPSS Inc., Chicago, IL) and analyzed using RDSAT 5.6, which provided weights controlling for selective recruitment bias and social network size. Adjusted frequency distributions were calculated with 95% confidence intervals.
The majority of FSW was younger than 30 years. Almost half of the women had less than 10 years of education. Nightclubs represented the most common place for meeting clients, followed by saloons, and streets. The work shift was predominantly nocturnal. Half of the women reported more than 7 sexual partners (including both steady partners and clients) during the week preceding the interview, and one third had 2 or more clients during the last day of work. The amount earned in the last day of work ranged from $7 to $750 US. One third of FSW reported their first sexual encounter between the ages of 7 and 14 years. Almost 90% of women reported regular use of condoms with clients in the previous month. Conversely, 20.9% of FSW regularly used condoms with their steady partners during the previous year (Table 1). More than two thirds of women had tattooing and/or piercings.
Some FSW reported more than one site of work.
RDS, respondent-driven sample; CI, confidence interval.
Alcohol consumption seemed to be a common behavior among these FSW. One third admitted drinking alcoholic beverages every day (at least 50 grams of alcohol). Use of illicit drugs included cocaine (12.9%) and marijuana (7.6%), and 9.3% used both. Only 26.7% of FSW obtained condoms from public health services. Most (35.9%) reported buying condoms with their own resources, 22.4% obtained them from their place of work, and 14.3% were provided with condoms by their clients. Half of the enrolled women had a previous history of vaginal discharge and almost 10% had genital ulcers or wounds during the previous 12 months (Table 1).
Six FSW (1.8%) were positive for anti-HIV-1/-2 antibodies and for HIV-1 RNA, and isolates from 4 FSW had the HIV-1 PR and RT regions sequenced. Two of 4 sequences analyzed were assigned subtype B based on PR and RT sequences. Two isolates had discordant HIV-1 in the PR/RT regions, and the B/F1 intersubtype recombination was confirmed by bootscanning analysis. No transmitted drug-resistant mutation was found among these newly identified isolates.
In the present study we found an HIV-1 infection prevalence of 1.8% in FSW that represented a 4.5- and 20-fold higher incidence rate compared to the 0.4% prevalence observed for the Brazilian female population in general 2 and the 0.09% observed in pregnant women in Goiânia City (0.09%), 8 respectively.
A wide variation in HIV-1 prevalence rates among female sex workers has been found worldwide. 3,4 Differences in regional endemicity, risk behaviors, and study designs probably are the reasons for these differences in observed rates. A recent systematic review reported by Malta et al. 10 showed a combined HIV-1 prevalence of 5.1% in Brazilian FSW; and Damacena et al., 3 (also using RDS) showed a global HIV-1 prevalence of 4.8% in 10 Brazilian cities. These rates are almost 3 times higher than that found in the current study. However, it is noteworthy that most studies analyzed by Malta et al. 10 were conducted in southeastern and southern cities, i.e., within or close to the epicenter of the Brazilian epidemic. The Damacena et al. 3 multicenter study did not provide rates for the different regions, including our study area.
Interestingly the HIV-1 prevalence found among FSW from Goiânia City was similar to rates found elsewhere in South America. An investigation of 13,600 FSW from 9 countries showed a global prevalence of 1.2%, ranging from 1% to 2% in the Andean (Venezuela, Columbia, Ecuador, Peru, Bolivia) and Southern Cone (Chile, Argentina, Uruguay, and Paraguay) regions, respectively. Only in two countries was the HIV-1 prevalence rate among FSWs above 2% (Paraguay, 2.6% and Argentina, 4.5%). 4
In the present study only one fourth of women were street-based sex workers, however, it was this subpopulation from which 5 of the 6 HIV-1–positive cases were identified. Women who recruited their clients on the street have been previously demonstrated to be at higher risk of acquiring an HIV-1 infection. 11 In fact, in the present study, these women were older, had lower education levels, served more clients per day, and their earnings were lower when compared with those who met clients in other places (data not shown).
The HIV-1 epidemic is complex and is affected by several factors, including individual behaviors, socio-cultural environments, and public health policies. In Goiânia City/central-western Brazil most FSW reported behaviors associated with protection against acquiring HIV-1 infection such as having at least 10 years of formal education, condom use during sexual intercourse with clients, and infrequent use of intravenous illicit drugs. However, this population had different risk factors associated with acquiring HIV-1 infections, including sexual encounters early in life, not regularly using condoms with steady partners, and obtaining condoms with their own resources (or from their clients), indicating low access to health services.
As observed by other studies, many FSW reported symptoms suggestive of previous STD (vaginal discharge). Even taking into account the limitation of this variable in predicting these infections, this finding suggested the potential risk of acquiring HIV-1 infections from FSW in Goiânia City. In fact, most FSW studied reported inconsistent condom use with their steady sexual partners, thereby creating opportunities for spreading STD including HIV/AIDS to uninfected populations.
We found 2 FSW infected with HIV-1 subtype B and 2 with recombinant mosaic BPR/F1RT isolates. All FSW were antiretroviral naïve and no pol gene sequences from these isolates had transmitted drug resistance mutations in contrast to a multicenter study conducted with men who have sex with men whose isolates demonstrated high resistance rates. 7 Furthermore, although a relatively small sample size was investigated, this study confirmed the spread and circulation of HIV-1 B/F1 mosaic forms among a heterosexual population in the Goiás State (central-western Brazil). 6,9,12
Despite the small sample size from the current study, the absence of transmitted resistance mutations contrasted with the 10% prevalence of transmitted drug resistant strains found following the analysis of 97 sequences isolated from naïve patients recruited in the same setting between 2007–2008, representing other vulnerable populations such as men who have sex with men. 9 Another study of 142 pregnant women infected with HIV-1 from the same geographical region analyzed PR/RT fragments of 83 HIV-1 isolates and identified 1 case of transmitted drug resistance among 17 naive women and 10/66 cases of secondary drug resistance. 12
The present study has limitations that should be taken account. We used an RDS sampling method with high rates of participation. This approach has been used to study vulnerable groups such as sex workers, illicit drug users, and MSM. Using this sampling method, we were able to study FSW in a range of settings including nightclubs, saloons, brothels, erotic movie houses, and on the streets. However, as in other studies, interpretation of these results and their application to other comparable populations and regions should be done with caution. Furthermore, interviewing in person may limit accurate responses to personal questions, such as risky sexual behavior. The small sample size is also a limitation for comparisons.
The present investigation provided data on HIV-1 epidemiology in a core group living far from the Brazilian HIV-1 epidemic epicenter. These data suggested that this population should be targeted for HIV-1 control and prevention programs among vulnerable women in Brazil. Since the rise of HIV-1 prevalence in sex workers increases the probability of spreading the epidemic, surveillance of HIV-1 prevalence in this vulnerable group should be conducted in a consistent basis in Brazil.
Footnotes
Acknowledgments
This research was supported by the Fundação de Amparo a Pesquisa do Estado de Goiás - FAPEG (CH 02/2008).
Author Disclosure Statement
No competing financial interests exist.