Article Text
Abstract
Objectives The aim of this study was to determine the prevalence of Chlamydia trachomatis infection and other sexually transmitted infections (STI) in female sex workers (FSW) in Tunisia.
Methods 188 prostitutes from three Tunisian towns were enrolled at their weekly medical visit. Demographic and sexual behaviour data were collected. C trachomatis, Neisseria gonorrhoeae, herpes simplex virus 2 (HSV-2) and human papillomavirus (HPV) were detected by PCR. Blood samples were tested for the presence of HIV, hepatitis B core, hepatitis C virus (HCV), HSV-2, C trachomatis and syphilis antibodies and Hbs antigen.
Results The mean age of the FSW was 34 years. They had worked in the sex industry for 6.6 years on average. Nearly all FSW (98.9%) had at least one marker of STI. A current infection was found in 86.7% of cases. Only one STI was noted in 37.2% and two or more in 49.5% of FSW. C trachomatis, N gonorrhoeae, HPV and HSV-2 PCR were positive in 72.9%, 11.2%, 44.1% and 1.1% of cases, respectively. Syphilis, HCV antibodies and Hbs antigen were detected in poor percentages, 2.7%, 1.1% and 0.5% of cases, respectively. No case of HIV infection was noted. No epidemiological or clinical factors were associated with STI. Only disturbed bacterial vaginal flora was found to be associated with C trachomatis infection.
Conclusion In this study, a high rate of C trachomatis infection was observed. The detection of this microorganism should be introduced in systematic surveillance of Tunisian FSW.
- Chlamydia trachomatis
- prostitution
- sexually transmitted infection
- sexually transmitted disease
Statistics from Altmetric.com
Sexually transmitted infections (STI) represent a major problem for public health in terms of morbidity and complications both in developed and developing countries. WHO estimates that more than 340 million new cases of sexually transmitted bacterial and protozoal infections occur throughout the world every year.1 Chlamydia trachomatis is responsible for the most common curable bacterial STI worldwide even in developed countries. This high prevalence of C trachomatis infections is aggravated by their potential contribution to facilitate HIV transmission. Besides, up to 70% of women and 50% of men with C trachomatis infections do not show symptoms of the disease and so are untreated. Consequently, the infection can become chronic and result in ectopic pregnancy, pelvic inflammatory disease and infertility.2 Individuals unaware of their infection increase the risk of transmission of C trachomatis in extramarital sexual contacts. Female sex workers (FSW) have been identified as one of the highest risk groups for the infection and transmission of STI because they are characterised by a high number of partners and poor healthcare-seeking behaviour.3 In our country, prostitution in venues is not allowed. Therefore, FSW work in brothels that are located in a district depending on the municipality. Legislation requires prostitutes to undergo medical control twice a week to be tested only for Neisseria gonorrhoeae. Blood samples are collected 3-monthly and tested for HIV, hepatitis B virus (HBV), hepatitis C virus (HCV) and syphilis. The only published study that evaluated several STI markers in this population in our country was conducted in 1988. Up to 70% of prostitutes had serological markers of C trachomatis infection.4 As a result of the lack of a national programme for preventing and screening for C trachomatis infection, we undertook this study in a large number of Tunisian FSW to determine the prevalence of C trachomatis infection and also other STI, and to assess the relationship between behavioural, clinical and microbiological factors and C trachomatis infection.
Materials and methods
Study population, data and specimen collection
The study was conducted by the laboratory of microbiology, Habib Bourguiba University Hospital of Sfax, in collaboration with hygiene clinics and doctors responsible for the medical control of FSW, in August/September 2007. The overall number of FSW in Tunisia is estimated to be 250. We conducted the study in three towns (Tunis, Sousse, Gabes) where the number of FSW was 188. FSW were checked at the same time as their weekly medical visit. A questionnaire was administered to each subject by face-to-face interview. It included questions on age, age when starting in commercial sex work, educational level, tobacco and alcohol consumption, drug use, number of instances of sexual intercourse per day, condom use and STI history.
For each FSW, a blood sample and two endocervical swabs were obtained. One swab was in alginate. The other swab, of Bactopick type, was eluted in 2SP medium.
All the subjects provided verbal informed consent, and the study protocol was approved by our ethics committee (Association d'Enregistrement et de Lutte Contre le Cancer du Sud Tunisien).
Laboratory testing
Standard bacteriological methods
The swab in alginate was subjected to Gram staining and bacterial culture on blood agar, chocolate agar–isovitalex 1% with and without vancomycin, colimycin, nystatin for N gonorrhoeae and culture of other common pathogens. N gonorrhoeae was identified by routine methods.Vaginal flora was evaluated with the Nugent score.5 When the score was 3 or less vaginal flora was normal, when between 4 and 6 vaginal flora was intermediate and when 7 or over the diagnosis of bacterial vaginosis was made.
Molecular methods
Endocervical specimens, eluted in 2SP, were tested for N gonorrhoeae and C trachomatis nucleic acids with the Amplicor CT/NG PCR-based assay (Roche Molecular Systems, Mannheim, Germany).
For the herpes simplex virus type 2 (HSV-2) and human papillomavirus (HPV) PCR tests, the extraction of the DNA was carried out by the Qiamp Mini kit (Qiagen GMBH, Hilden, Germany). The HSV2 and HPV DNA amplification and detection as well as the characterisation of the HPV oncogenic groups were carried out by Hybridowell universal primers and probes for HSV-2 kit and HPV consensus kit (Argène, Varhiles, France), respectively.
Serological testing
Chlamydial IgG antibodies were determined using simplified method of the microimmunofluorescence assay. The antigens of three Chlamydiae species have been used: the L2 C trachomatis strain, the Loth Chlamydia psittaci strain and the IOL-207 Chlamydophila pneumoniae strain. The microimmunofluorescence assay was performed as previously described.6 The cut-off titre for positive sera was 1/32 for C trachomatis IgG antibodies.
Blood samples were also tested for syphilis antibodies using the Treponema pallidum haemagglutination assay (BioMaghreb, Tunis, Tunisia).
The HSV-2 IgG antibodies were detected by commercial ELISA (Vircell, Granada, Spain). Hbs antigens and antibodies to hepatitis B core antigen were detected by commercial ELISA kits: The HBsAg Confirmatory Test and the ETI-AB-Corek Plus (DiaSorin, Saluggia, Italy), respectively, according to the manufacturers' instructions. Screening for the HIV P24 antigen and the antibodies to HIV-1 and HIV-2 in FSW sera was performed using the Genscreen ultra HIV Ag–Ab test (Bio-Rad, Marne-la-Coquette, France).
Statistical analysis
Data analysis was performed using SPSS version 13.0. The prevalence of each STI was calculated with the 95% CI. The relationship between sociodemographic characteristics, sexual behaviour, clinic data, microbiological data and C trachomatis infection was examined by contingency tables. Differences were statistically compared using the χ2 test; p<0.05 was considered significant.
Results
Demographic characteristics
All 188 FSW invited to participate consented. Blood samples were collected from 183 women and endocervical specimens were obtained from all participants. Sociodemographic data and sexual behaviour of the Tunisian FSW are shown in table 1. The mean age of participants was 34 years. Thirty-two FSW (17%) had never attended school, 87 (46.3%) had attended primary school, 65 (44.6%) secondary school and four (2.1%) university. The mean duration of commercial sex work was 6.6 years ranging from 1 month to 20 years. Sixty-three (33.5%) had worked as a prostitute for less than 5 years and were characterised as new FSW. The latter were younger, consumed less tobacco (74.6% vs 87.2, p=0.03) and were more active because they had a significantly higher mean number of clients per day (30.8 vs 24.3, p=0.01). New FSW were better schooled and had a greater tendency to use condoms. Frequent condom use was noted in 73% of the new FSW and in 54.4% of the old FSW (p=0.01, OR 0.44).
Clinical data
Clinical findings of the Tunisian FSW are shown in table 2. A history of genital tract infection was found in 79 cases (42%). These infections more frequently affected the lower genital tract (33.5%). Old FSW more frequently reported a history of upper genital tract infection than the new FSW (11.2% vs 3.2%; p=0.05). They also more frequently reported a history of gynaecological surgery (44.8% vs 23.8%; p=0.005). At the time of our visit, leucorrhoea was noted significantly more frequently in the new rather than in the old FSW (9.5% vs 2.4%; p=0.04). Exocervicitis was observed in 4.8% of cases with no statistical difference between new and old FSW.
Microbiological data
The prevalence rates of different pathogens detected in Tunisian FSW are shown in table 3. Among the microorganisms tested for, any marker of STI (positive PCR and/or serology) was detected in nearly all FSW (98.9%) demonstrating that they had a current STI or had contracted a STI at least once in their life. A current STI was detected in 86.7%, only one STI in 37.2% and two or more STI in 49.5% of FSW. When the vaginal flora was evaluated with the Nugent score, it was normal in only 19.7% and intermediate in 47.3%. Bacterial vaginosis was noted in 33% of cases. C trachomatis was the pathogen most frequently detected. In fact, PCR was positive in 72.9% of cases and serology in 85.8% of cases. When we compared new with old FSW, no statistical difference was noted in C trachomatis PCR positivity; however, the latter had significantly higher C trachomatis IgG titres (≥128) than the former (p=0.02). N gonorrhoeae was detected in only seven cases (3.7%) by culture and in 21 cases (11.2%) by PCR. The prevalence of syphilis was 2.7%. HSV-2 PCR was positive in only two cases (1.1%). In contrast, a high seroprevalence of HSV-2 (55.5%) was noted in this population. HPV DNA was detected by PCR in 83 cases (44.1%). Genotyping was possible in 51 of the cases. High-risk oncogenic HPV types were detected in 43 cases (84.4%), whereas low-risk oncogenic HPV types were detected in four cases (7.8%). In the four remaining cases (7.8%), the two types of HPV were found. Although Hbs antigen was detected in only one case (0.5%), hepatitis B core IgG antibodies were present in 59 cases (31.4%). They were more frequently detected among the old FSW (p<0.001). HCV serology was positive in only two cases. None of Tunisian prostitutes was HIV positive.
The univariate analysis of the risk factors for STI and C trachomatis infection showed no association between any demographic or behavioural factor and any STI or C trachomatis infection (table 4). However, FSW with C trachomatis-positive PCR had more altered vaginal flora at the Gram staining than those not infected or infected with other microorganisms (p=0.004).
Discussion
This study provides new information about STI and sexual behaviours in Tunisian FSW. In fact, the last study was conducted 20 years ago and was limited to only one town (Sousse).4 Therefore, our study is the largest to be carried out in Tunisia and the first to screen for many STI pathogens in FSW. We aimed to determine in Tunisian prostitutes the prevalence of most reported STI: C trachomatis, N gonorrhoeae, syphilis, HSV-2, HPV, HBV, HCV and HIV. Detection of Trichomonas vaginalis was not performed in our study because of condition limitations (samples must be transported from consultation clinics to the laboratory). We confirmed that the STI rates are high among FSW in spite of the absence of HIV seropositivity. The prevalence of any STI in FSW, found in our study, was higher than those reported in the literature. This prevalence ranged from 6.5% in North America7 to 84% in south Asia.8 In 2005, it was 24.4% in Cambodia,9 33.6% in Mexico10 and 79% in Yunnan in China.11 The high percentages reported in Africa are related to the high prevalence of HIV infection, whereas in Asia, trichomoniasis and C trachomatis infections are the most prevalent STI. In this study, this figure, in which the C trachomatis infection rate is considerably high, is far from other reports in the world.
The mean age of our FSW is higher than those reported in other studies.10–13 In Tunisia and for social considerations, women working in brothels could not have a normal family life, so that they did not abandon the brothel. Moreover, Tunisian legislation does not allow venue sex work. FSW must work in brothels with a fixed earning per client, with policy permission and under the control of municipal health services.
Condoms were used frequently by 60.6% of FSW and sometimes by 39.4% of them. Condom use varies within locations worldwide. In China, for example, condoms were reported not to be used by 2.7–15.5% of FSW, to be used sometimes by 38.8–50.7% and to be used frequently by 38.9–58.8% of cases.14 In Barcelona, condom use was reported by 96% of female street prostitutes.15 Our FSW had free access to condoms, which are distributed by social health clinics. However, regarding the low acceptability of their clients, the percentage of condom use reported in our study seems to be overestimated. In our study, frequent condom use was, significantly, reported more by new FSW. The latter appear to have better educational levels, which may lead to better awareness of STI prevention methods.
The high number of clients per day reported by our FSW is probably related to the period of enrolment being in the summer. In fact, students visit brothels more frequently in the holidays. Moreover, tourism from internal regions as well as neighbouring countries is more pronounced in this season. Activity was very busy even in the morning, at the time of our visit.
Our study showed that gynaecological problems are frequent among FSW. In fact, 33.5% had contracted at least once a lower genital tract infection. We also noted that this population is vulnerable to upper genital tract infection and gynaecological surgery. These complications appeared, significantly, more frequently in FSW working for more than 5 years. Leucorrhoea was more frequently noted among new FSW. In fact, the latter are probably less habituated to this symptom and automedication in this group is less frequent than in old FSW.
Our study confirmed that rates of STI were high among FSW in Tunisia and showed that C trachomatis was the most common STI, followed by HPV, then N gonorrhoeae. C trachomatis infection seems to be prevalent in our country. In fact, the prevalence of the bacterium was 43.3% in infertile men16 and 59% among patients with arthritis in our region.17 This high prevalence is far from what is reported in the literature. The prevalence of C trachomatis infection reported among FSW is very low in developed countries and high in developing countries, but rarely higher than 50%. In Spain, C trachomatis was detected in 4.7% of FSW in Barcelona and 5.9% in Catalonia.12 15 In Italy, the C trachomatis infection rate among FSW was 14%.18 In Asian countries, this prevalence ranged from 14.4% in Cambodia9 to 46.3% in China.19 In Latin America, a recent study conducted in Mexico showed that C trachomatis infection prevalence was 16.9%.10 In African countries, rates reported were up to 24.5% in Madagascar.20 To detect C trachomatis on endocervical swabs, the Cobas Amplicor PCR was used in our study. This automated PCR test has been reported to be sensitive and specific up to 99%, and the risk of contamination with this test is very low.21 Besides, 85.8% of FSW had positive serology showing that C trachomatis infection is very frequent in these women. However, in a recent review, the reproducibility of PCR has been questioned. A positive PCR result was reported not always to be associated with current infection.22
HSV-2 seroprevalence in our FSW was also high (55.5%). Similar or higher rates were also reported worldwide, either in African countries, Asia, and developed countries such as in the USA, where it was 72.9% among low income women who reported a history of sex work.23–25 In spite of this high HSV-2 seroprevalence, we did not note any genital ulcers. In line with our findings, most infections caused by HSV-2 were reported to be asymptomatic, with intermittent genital excretion of the virus, although it could cause severe and recurrent anogenital vesicles and ulcers.26 Transmission of HSV-2 to a sexual partner occurs within the whole clinical spectrum of this infection even during periods of asymptomatic virus excretion.27 The prevalence of cervical HPV infection found in our study is similar to previous reports in our country. Cervical HPV was thus found in 39.2% of prostitutes from Sousse and in 43.7% of those from Tunis.28 29 The carriage of high oncogenic risk HPV types was also reported in these studies. In fact, FSW are mostly exposed because they have multiple sexual partners. Other viral STI tested in our study had a low prevalence. In fact, HCV and HBV were found in only 1% and 0.5% of cases, respectively. The HIV infection rate was 0%, because seropositive FSW were automatically excluded from the exercise.
New FSW more frequently had two STI than old FSW, but no statistical difference was observed (p=0.13). Furthermore, the prevalence of bacterial STI was slightly higher among new FSW in spite of their frequent condom use. In fact, new FSW are more active and have a significantly higher mean number of clients per day than old FSW. N gonorrhoeae was detected in 17.5% among the new FSW compared with 8% among old FSW (p=0.05). C trachomatis prevalence was slightly higher among new FSW (76.2% vs 71.2%, respectively). However, old FSW had, significantly, more markers of previous infections. Effectively, high titres of C trachomatis IgG antibodies were detected much more frequently among old FSW (p=0.02). Similarly, the prevalence of IgG antibodies to hepatitis B core antigen was much higher among old FSW (p<0.001). These data demonstrate that the more they work in the sex industry, the more they are exposed to STI.
No behavioural factor was found to be associated with any STI. This is due to the high prevalence of STI, the same conditions of work of FSW, and essentially the fact that they had the same type of clients. In our study, we found that women with chlamydial infection less often had a normal vaginal flora at Gram staining (p=0.004). That was in line with a recent study that showed an association between bacterial vaginosis and C trachomatis cervicitis.30 These results indicate that when the bacterial flora is altered, the woman is at a high risk of having a STI, especially C trachomatis, and should be tested for this infection.
In conclusion, our cross-sectional study confirmed that the prevalence of STI is high among FSW in Tunisia. In spite of the absence of HIV infection in women working in the sex industry, a high rate of C trachomatis infection was observed. This high prevalence indicates frequent risky sexual practices and a poor provision and uptake services. In fact, for controlled STI, low rates of prevalence were observed. We propose, primarily, to treat all FSW, and secondarily to promote surveillance programmes by introducing C trachomatis diagnostics in the panel of STI controlled for in our FSW.
Key messages
Any STI marker (positive PCR and/or serology) was detected in nearly all FSW (98.9%).
C trachomatis was the most frequently detected pathogen with positive PCR in 72.9% of cases and positive serology in 85.8% of cases.
None of our FSW had positive HIV serology.
Detection of C trachomatis should be introduced in systematic surveillance of Tunisian FSW.
Acknowledgments
The authors would like to thank Drs R Ben Mahmoud from Tunis, H Knani from Sousse and M Ben Othmen from Gabes, for their collaboration.
References
Footnotes
Funding This study was funded by the Family Planning ‘Office National de la Famille et des Population’ and the research laboratory ‘Microorganismes et Pathologie Humaine’. de Tunis, Laboratoire de Recherche MPH.
Competing interests None.
Patient consent Obtained.
Ethics approval This study was conducted with the approval of the ethics committee of the Association d'Enregistrement et de Lutte Contre le Cancer du Sud Tunisien.
Provenance and peer review Not commissioned; externally peer reviewed.