Prevalence and factors associated with Trichomonas vaginalis infection among men who have sex with men and female sex workers in Togo, 2017.

Background The aim of this study was to estimate the prevalence and factors associated with Trichomonas vaginalis ( T. vaginalis ) among men who have sex with men (MSM) and female sex workers (FSW) in Togo in 2017. A cross-sectional bio-behavioral study was conducted from August to October 2017 using a respondent-driven sampling method in four cities in Togo. A standardized questionnaire was used to record socio-demographic data and sexual behavior patterns. T. vaginalis detection by molecular biology tests was performed using Allplex STI Essential Assay which detect also 6 others micro-organisms. A blood sample was drawn and serological test using SD Bioline Duo VIH/Syphilis rapid test was performed for HIV and syphilis testing. of T. vaginalis in these populations and in other populations in Togo.


Introduction
Trichomonas vaginalis (T. vaginalis) is the most common, curable parasitic sexually transmitted infection (STI) worldwide affecting both men and women [1]. In 2012, 143 million cases of T. vaginalis had been diagnosed in women aged 15-49 years worldwide, including 17.5 million in Africa [2]. In 2016, Bayesian meta-analysis was used to generate estimates of the prevalence of STI. In women, prevalence estimates for T. vaginalis was 5.3%, for Chlamydia Trachomatis (C. trachomatis) 3.8%, for Neisseria gonorrhoeae (N. gonorrhoeae) 0.9%, and for syphilis 0.5%. In men, prevalence estimates for T. vaginalis was 0.6% which was very low compared to C. trachomatis (2.7%) and N. gonorrhoeae ( 0.7%) [3]. T. vaginalis vaginal infection in the African region are estimated at 42.8 million, and in the same region, this infection is ten times more common in women than in men [4]. As most STI, T. vaginalis infection is largely associated with an increased risk in HIV acquisition.
To our knowledge, no study on T. vaginalis infection using molecular technique has been conducted in Togo, especially among key populations. The objective of this study was to estimate the prevalence and factors associated with T. vaginalis among key populations in Togo.

Study design and recruitment
This study was a bio-behavioral cross-sectional study conducted among MSM and FSW from August to October 2017 in four cities of Togo: Lomé, the capital city, Kpalimé, Atakpamé, and Tsévié. Togo is a West African country with 7.6 million inhabitants in 2018, covering 57,000 square kilometers with HIV prevalence in general population estimated at 2.3% in 2018 [20].
Participants were included through the respondent-driven sampling method [21,22]. MSM were defined as men who have had sex with other men within 12 months prior to recruitment, and FSW were defined as women having had sex in exchange for money as a compensation in the previous 12 months. Additional inclusion criteria were being age ≥ 18 years, living in Togo more than 3 months and having given written informed consent.
Study procedures and detection of T. vaginalis and other STI A standardized questionnaire adapted from a Family Health International (FHI) 360 validated guide for bio-behavioral surveys was administered during a face-to face interview to collect information regarding socio-demographic characteristics, risky sexual behaviors.

Statistical analysis
Descriptive analyses were performed and results were presented with frequency tabulations and percentages. Prevalence were estimated with their 95% confidence interval (95%CI). Chi-square or Fisher's exact tests were used to compare categorical variables. In multivariable analysis, logistic regression was conducted to identify factors associated with T. vaginalis infection.
Associations in the regression model were expressed as adjusted odds ratio (AOR) using all variables that had p < 0.2 in the univariable regression. Predictor variables were selected as those found to be relevant according to the literature review. All computations were conducted using R© version 3.4.3 software and the level of significance was set at 5%.  Table 1.

Discussion
This study provided an update on the epidemiology of T. vaginalis infection and showed the absence of the infection among MSM (0.0%) and a low prevalence among FSW (6.5%) in Togo. The overall prevalence of other STI were 16.8%, 9.3%, 7.5% and 7.2% for HIV, M. genitalium, C. Trachomatis and N. gonorrhoeae infections respectively. Among FSW population, risk factors associated with T.
vaginalis infection were the geographic area (capital city, Lomé), lower age at first intercourse and infection with C. Trachomatis.
Our study reported that prevalence of T. vaginalis infection was 6.5% among FSW. In Rwanda, a descriptive cross-sectional study conducted in 2015 among 1,168 FSW reported a prevalence of 11.9% [23]. In a 2-year longitudinal study conducted among 350 Kenyan FSW, baseline prevalence of T. vaginalis was 9.2% [24]. In an another prospective cohort study among 352 South African youths including lesbian, gay, bisexual, transgender, and queer (LGBTQ), an overall prevalence of 4.8% has been reported (8.1% among female and 0.7% among male participants) [25]. In a prospective, interventional cohort study of FSW aged 18 to 25 years in Ouagadougou among 321 HIV-uninfected FSW the prevalence of T. vaginalis was 3% [26]. A prospective study among 302 pregnant women conducted in 2011 in Togo reported a prevalence of T. vaginalis of 3.7% [27]. In a cross sectional study conducted in Mexico in 2011 among 105 FSW, the prevalence of T. vaginalis was 25.7% [28] which contrasts with that reported in FSWs in our study (6.5%). This difference could be explained by the methods used for the diagnosis of T. vaginalis, the age of the population and the associated risk factors.
Concerning associated risk factors among FSW, as observed in our study, C. trachomatis infection vaginalis infection in Kenya [24]. Also, in the same study, a significant association was reported vaginalis infection among adolescents and young adults who were younger at the time of their first sexual intercourse [30].
Also consistent with our result, a cross-sectional study conducted in four cities in sub Saharan Africa (Kisumu, Kenya; Ndola, Zambia; Cotonou, Benin and Yaoundé, Cameroon) among a random sample of 8,000 adults (2,000 in each city), aged 15-49 years showed a prevalence of T. vaginalis respectively of 29.3% in Kisumu, 34.3% in Ndola, 3.2% in Cotonou and 17.6% in Yaoundé. Early sexual debut (before age 15) was a significantly risk factor associated with T. vaginalis infection in women in Ndola (Zambia) [31].
In our study, no T. vaginalis was detected in MSM, which is different from prevalence found in similar population in African countries: 2,1% reported in Côte d'Ivoire in 2008 [32] and 9% in South Africa in 2018 [33]. In the Netherlands in 2014, the overall prevalence of T. vaginalis infection among 1,204 heterosexual men and MSM was respectively 1.1% and 0.0% [34], nearing our results. Reasons of prevalence disparities between FSW and MSM are not clear and the hypotheses are not confirmed.

One of the most likely hypotheses is that T. vaginalis probably does not develop in the rectum and is
therefore not often present in MSM [35]. To our knowledge, this was the first study reporting prevalence of T. vaginalis infection and STI among MSM and FSW in Togo. Another strength of this study includes the use of a sensitive laboratory assay for the reliable detection of T. vaginalis infection and the relationship with HIV. Finally, our study which was the first assessing factors associated with T. vaginalis among FSW in Togo, also provided useful information in order to design specific interventions within these populations.
There were few limitations to this study including the lack of data on treatment use among study participants, which may have certainly impacted observed STI prevalence. Furthermore, the standardized questionnaire submitted to participants can be biased (memory bias and social desirability bias) by the fact that it was based on self-reporting and may not reflect the overall sexual activity. Additionally, due to the cross-sectional nature of this analysis, we are unable to analyze the causality and temporality of the associations between T. vaginalis infection and other factors. Finally, because T. vaginalis infection was null among MSM in the study, we were unable to assess the relationship of T. vaginalis infection and predictor variables among this population.

Conclusion
The prevalence of T. vaginalis infection using molecular test among MSM and FSW in Togo was null and low, respectively. However, extensive studies are needed to confirm and better understand the

Availability of data and materials
All data generated or analyzed during this study are included in this published article [and its supplementary information files].

Competing interests
The authors declare that they have no competing interests