Article Text

Download PDFPDF

Prevalence of Trichomonas vaginalis infection in women attending a major gynaecological hospital in Greece: a cross-sectional study
  1. Evangelia-Theophano Piperaki1,
  2. Marianna Theodora2,
  3. Michael Mendris3,
  4. Louisa Barbitsa3,
  5. Vassiliki Pitiriga1,
  6. Aris Antsaklis2,
  7. Athanassios Tsakris1
  1. 1Department of Microbiology, Medical School, University of Athens, Athens, Greece
  2. 2First Department of Obstetrics and Gynecology, Alexandra University Hospital, Athens, Greece
  3. 3Clinical Microbiology Laboratory, Alexandra University Hospital, Athens, Greece
  1. Correspondence to Professor Athanassios Tsakris, Department of Microbiology, Medical School, University of Athens, 75 Mikras Asias Street, 11527 Athens, Greece; atsakris{at}


Background The prevalence of Trichomonas vaginalis is not accurately estimated, since it is not a reportable disease.

Aims To assess the prevalence of T vaginalis infection in women attending a Greek gynaecological hospital and to evaluate four diagnostic methods for T vaginalis infection.

Methods 255 symptomatic and 247 asymptomatic women were included in the study during 2006–07; 372 were Greek and 130 were immigrants. T vaginalis was detected in vaginal samples, using wet mount, culture in modified Diamond's medium, antigen detection and two PCR assays, targeting different regions of T vaginalis genome. Specimens were considered positive for T vaginalis, when tested positive either by culture or by both PCRs.

Results 23 women (4.6%) were positive for T vaginalis. Seven of the 23 positive samples (30.4%) were only PCR-positive. Infection was more prevalent in symptomatic women (6.7%) than in asymptomatic ones (2.4%). T vaginalis was more frequently detected in immigrants (7.9%) than in Greek women (3.3%). Gardnerella vaginalis infection was significantly more frequent in women infected with T vaginalis. PCR was the most sensitive method (100%), followed by culture (69.6%), wet mount (69.6%) and latex agglutination (54.6%). Agreement between PCR and culture as well as wet mount examination was very good (κ=0.79).

Conclusions The study shows a relatively low percentage of trichomoniasis in the female population living in Athens. The infection was more prevalent among immigrants, and a proportion of the infected women was asymptomatic. The tested methods had good agreement and PCR was found to improve the diagnostic yield considerably.

  • Trichomonas
  • Greece
  • PCR
  • culture
  • latex

Statistics from


Trichomonas vaginalis is a flagellate protozoan parasite that causes trichomoniasis, one of the commonest sexually transmitted infections. Humans are the only known host and the parasite is transmitted sexually.1 T vaginalis causes asymptomatic, acute or chronic urogenital infection in women, ranging from acute vulvovaginitis to mild chronic vaginitis with pruritus, dyspareunia and scanty vaginal secretion.2 As much as 50% of the infected women may be asymptomatic carriers, although half of those develop symptoms within 6 months.3 Trichomoniasis is much less frequent in men, who may be asymptomatic carriers or suffer acute or mild urethritis.4 T vaginalis has a widespread geographical distribution and is highly prevalent. The World Health Organization estimates that 170 million new cases occur each year.5 The highest rates are reported from developing countries,6 from high risk populations, such as sex workers7 and from socioeconomically disadvantaged populations, living in developed countries.8 However, exact numbers are difficult to obtain because the infection is not reportable and prevalence is also generally underestimated, due to the poor sensitivity of routinely employed diagnostic tests.

Diagnosis of T vaginalis infection in most parts of the world, including Greece, is based on microscopic examination of a saline wet mount. A positive wet mount is highly specific,9 although a negative result cannot exclude the diagnosis, due to its low sensitivity, which ranges from 30% to 80%.10 11 Both the sensitivity and the specificity of wet mount depend greatly on microscopist experience. Culture is considered the gold standard for T vaginalis diagnosis, with an overall sensitivity of up to 95%,12 but requires as long as 7 days for results to be obtained. PCR based assays for the diagnosis of trichomoniasis were introduced in the early 1990s13 and have since aided the diagnosis of the infection, both in women and men, with an overall sensitivity and specificity around 95% and 98%, respectively.12

Trichomoniasis has been the focus of few specific studies in European countries and elsewhere. Information on its prevalence is usually derived from surveys of other sexually transmitted infections. In Greece, the synthesis of the population has changed significantly in the past two decades, due to the constant influx of large numbers of immigrants and refugees which has probably affected the epidemiology of several infectious diseases in our country, including sexually transmitted infections. A significant proportion of this population has access to the Greek national health system. The aim of the present study was to assess the prevalence of trichomoniasis in the population visiting a major gynaecological hospital in Athens, using different laboratory methods: culture, wet prep examination, latex agglutination test and two different PCR assays; also to compare the sensitivity and specificity and to determine the agreement between these methods. To our knowledge, this is one of few systematic cross-sectional studies, employing PCR for the determination of T vaginalis prevalence.

Materials and methods


Specimens were collected consecutively, during the period 2006–07, from 502 women attending the outpatient clinic of Alexandra Hospital, a major gynaecological hospital in Athens. Women with or without genital complaints were asked to participate in the study. Women without genital complaints were those attending the outpatient clinic for routine gynaecological tests (Papanicolaou screening test). Women with genital complaints were referred by their gynaecologists for laboratory investigation of their symptoms. All women completed a questionnaire, including demographic data, reproductive history, behavioural and sexual information, history of previous sexually transmitted diseases, prior antibiotic and other drug use, and the nature and duration of their current symptoms. The questionnaires were anonymous and linked to the participants by a code number. Informed consent was obtained from all study participants.

Participants were included in the study, according to the following criteria: intact uterus, no use of vaginal medication in the previous 3 days, no sexual intercourse during the previous 3 days. All participants were tested for the presence of T vaginalis, with four laboratory methods, using vaginal fluid samples. Data regarding the presence of other microorganisms, obtained by vaginal fluid culture, Gram stain and wet mount examination, was also recorded.

Specimen collection

Four vaginal fluid samples were collected from each participant, from the vaginal fornices and lateral walls, using sterile synthetic swabs. One swab was used to inoculate the liquid culture medium (Diamond's modified medium). A second swab was used for preparation of a saline wet mount. A third swab was used for antigen detection, using the Kalon TV latex agglutination test (Kalon Biological, Surrey, UK). A fourth swab was eluted in 500 μl saline and stored at −20°C for PCR.

Wet mount

The swab was placed in 0.9% saline; it was briefly agitated and a wet mount of the eluate was examined under 400× magnification, for the characteristic morphology and motility of T vaginalis.

Antigen detection

Antigen detection was carried out within 2 h of specimen collection, using the Kalon TV latex agglutination test (Kalon Biological), according to the manufacturer's instructions.

Vaginal fluid culture

Vaginal fluid cultures were performed in the hospital microbiology laboratory. Samples were cultured on blood agar, MacConkey and Sabouraud agar, and were incubated at 37°C aerobically and anaerobically for 48 h. Presence of Candida spp. was diagnosed by wet mount microscopy, Gram stain and/or culture. Gardnerella vaginalis was diagnosed, based on the presence of clue-cells in microscopy, on Gram stain morphology and/or aerobic and anaerobic culture.

Culture of T vaginalis

Culture was performed in liquid modified Diamond's medium (MDM).14 MDM was prepared every two weeks and stored at 4°C, until use. The vaginal swabs were placed in conical tubes containing 7 ml of MDM and mixed by gently vortexing to dislodge the cells. They were left to incubate for a total of 7 days, and were examined microscopically with a wet mount, on days 3, 5 and 7 after inoculation. A positive result was defined as the presence of motile trichomonads at any time, whereas a negative result was defined as the absence of motile trichomonads at all readings. Sterility of the medium was ascertained by culture on blood agar.

PCR assays

DNA isolation was performed using the Wizard Genomic DNA purification kit (Promega), according to the manufacturer's instructions. All samples were initially amplified using the BTUB 9/2 primer set, which is designed to target conserved regions of the three beta-tubulin genes of T vaginalis, as previously described.15 Positive samples were consequently amplified with the AP 65A/B primer set, which is designed to target the adhesin genes of T vaginalis.16 Different diagnostic tests were performed by different investigators to avoid bias. Specimens were considered positive for T vaginalis, when tested positive either by culture or, in the event of negative culture, by both PCRs, employing different primer sets, according to previously described criteria.15

Statistical analysis

Culture is considered to be the gold standard for the detection of T vaginalis. For our statistical analysis we have used an expanded gold standard, where specimens were considered positive for T vaginalis if they had a positive culture or/and two subsequent positive PCRs. Sensitivity and specificity values were compared among the laboratory tests and κ index for measurement of tests agreement was determined. The coexistence of T vaginalis with abnormal vaginal flora was also assessed. Categorical variables were analysed using the χ2 test. Continuous variables were analysed using the independent t-test. p Values of 0.05 or less were considered statistically significant. Statistical analysis was performed using the statistical package STATA V.9.0.


A total of 502 women were included in the study. Table 1 shows the characteristics of the study population. The mean age of the women was 39.7 (17–65 years, SD=12.4). The majority was of Greek origin (74.1%), married (66.5%) and had no history of trichomoniasis (93.1%) or other sexually transmitted infection (83.2%).

Table 1

Demographic and clinical characteristics of the study population (n=502)

On presentation, 255 (50.8%) had symptoms indicative of vaginitis, such as vaginal discharge (26.8%), pruritus (18.4%), dysuria (4%), dyspareunia (5.3%) or any combination of the above. The remaining 247 (49.2%) were asymptomatic (table 1). Women asymptomatic at presentation were significantly older than symptomatic ones (mean age 44.3, SD=12.2 and mean age 35.2, SD=10.8, respectively; p=0.04). Women with pruritus were significantly younger than those without (mean age 35.8, SD=12.0 and mean age 40.2, SD=12.3, respectively; p=0.003). Women with vaginal discharge were significantly younger than those without (mean age 35.6, SD=11.2 and mean age 40.7, SD=12.5, respectively; p=0.0001). Other symptoms (dysuria and dyspareunia) were not statistically associated with the age of the population.

Of the 502 women, 16 (69.6%) were found positive for T vaginalis by culture, 16 (69.6%) by wet mount and 12 (54.6%) by latex agglutination. It is worth noting that not all samples positive by wet mount were culture positive and vice versa. When the first PCR was applied, 24 specimens were positive, including all 16 culture positive samples and eight culture negative. Of the eight specimens, with discrepant results between culture and PCR, seven were determined to be true positives, when amplified by a second pair of primers, targeting a different region of the T vaginalis genome. Thus, applying the extended gold standard, a total of 23 (4.6%) positive samples for T vaginalis were identified. Seventeen of the positive samples belonged to symptomatic women and six to asymptomatic women, therefore the prevalence of trichomoniasis in the symptomatic group was 6.7% and in the asymptomatic group 2.4% (p=0.02). Three of the six asymptomatic women with T vaginalis infection were found positive by latex agglutination (50%), and five (83.3%) by culture. Table 2 shows a comparison of the sensitivities, specificities, and positive and negative predictive values of the employed diagnostic methods, compared to the expanded gold standard. PCR showed the highest specificity (100%). Wet mount and culture showed the same sensitivity (69.6%), whereas latex agglutination showed the lowest sensitivity (54.6%). κ Test for agreement between diagnostic tests was also determined (table 3). The κ index was 0.94 between culture and wet mount, 0.81 between culture and latex, and 0.79 between culture and PCR.

Table 2

Performance of all diagnostic tests (95% CI) (n=502)

Table 3

κ Index for agreement between diagnostic tests for T vaginalis

A total of 166 (33.1%) participants were positive for Candida spp. and 67 (13.4%) were positive for G vaginalis. Of the women with history of previous T vaginalis infection (n=35), none tested positive, whereas of those with history of other sexually transmitted infections (STIs) (n=84), only one (1.2%) was found T vaginalis-positive.

Table 4 presents a comparison of women with and without trichomoniasis. Women with trichomoniasis were more often symptomatic than those without trichomoniasis (p=0.02). A statistically significant difference was also observed regarding the nationality of infected women. Non-Greek women comprised 45.5% of the infected group and 24.9% of the uninfected group, whereas Greek women comprised 54.5% of the infected group and 75.1% of the uninfected group (p=0.04). G vaginalis was detected with a significantly higher frequency in women infected than in those not infected with T vaginalis, specifically in 11 women (57.9%) of the infected group and 16 (8.7%) in the non-infected group (p<0.001). Also, vaginal discharge was more often present in the infected group than in the non-infected group, specifically in 12 of the infected women (52.2%) and in 123 (25.7%) of the non-infected women (p=0.007). No statistically significant differences were found in age and other characteristics between the two groups.

Table 4

Characteristics of women infected and non-infected with T vaginalis


The prevalence of trichomoniasis in women varies greatly, depending on the population studied. In a general, low risk population, the prevalence of trichomoniasis is estimated to be less than 3.5%.17 In STI clinics and other specific populations however, the prevalence of this microorganism may vary between approximately 4.8% and 25%, depending on the specific characteristics of the population.18 Several studies have also determined a much higher prevalence of infection (10.0–18.5%) among young females living in urban areas.19 20 Recent European studies, performed mostly on specific populations, such as prison inmates and HIV seropositive individuals, have found frequencies ranging from approximately 8% to 30%.21–23 In Greece, there is limited data available regarding the prevalence of this sexually transmitted infection, since trichomoniasis is not a reportable disease, nor is it the focus of any public health control programmes, probably because of its mild nature. Approximately half of the women infected with T vaginalis are asymptomatic, as is the majority of infected men. Symptomatic women usually present with vaginal discharge, pruritus and irritation of varying severity. Dysuria or abdominal pain may be occasionally observed.1 In a Greek study of women with vaginal discharge, conducted in 1986, the prevalence of T vaginalis was approximately 20% by wet mount examination, and an association was found between trichomoniasis and education level.24 In a more recent study that included only patients with symptoms of vaginitis, the prevalence of T vaginalis was 8.1%.25 However, both of these studies were retrospective, examined exclusively women with vaginitis, and employed wet mount only, for the diagnosis of trichomoniasis. The present survey is a cross-sectional study in a population of Greek and non-Greek women, and used four different methods to ascertain parasite detection, including PCR assays with different molecular targets. The distinguishing feature of the present study, is that it has estimated T vaginalis prevalence in both symptomatic and asymptomatic women and also compared local women to immigrants.

The risk indicators for trichomoniasis were investigated, by comparing demographic and clinical characteristics between infected and non-infected women. No differences regarding age, previous STI or previous trichomoniasis were found, in contrast with other similar studies, which found trichomoniasis prevalence to increase with age.23 26 27 Recent findings, however, suggest that trichomoniasis, often asymptomatic, may be more common than previously thought in younger women.28

Immigrants were found to be infected more frequently than Greek women in the study population (p=0.04). The significant difference in nationality could be explained by the fact that the majority of non-Greek women included in this study originated from Balkan and East European countries and were living as immigrants in our country, under relatively poor socioeconomic conditions. Moreover, the existing differences in religious and social background, as well as sexual health education, between Greek and immigrant women, may have contributed to their personal hygiene standards and sexual practices, both of which are considered major risk factors for Trichomonas infection.

Infection with T vaginalis is generally a marker of high-risk sexual behaviour and its coexistence with other STIs is common.29 An association was found between T vaginalis and G vaginalis infection, consistent with the findings of other studies from various countries.30–32 Since the present study was cross-sectional, it was not possible to determine which of the infections occurred first, perhaps implying a causal role for either of the two; still, the frequent coexistence of the two infections warrants further investigation of the mechanisms underlying this association.

Wet mount preparation is most commonly used for the diagnosis of trichomoniasis, given its advantages such as speed and low cost,33 although it fails to detect all culture positive patients. Culture, although it requires complex culture media, an extended incubation period (up to 7 days) and daily monitoring of parasitic growth, is still a more reliable method. This, however, makes it difficult to assess the sensitivity of PCR, which is often reported higher than that of culture. In order to overcome this problem, an expanded gold standard was applied in the present study. In the event of negative culture and positive PCR, the samples were amplified using a second set of primer, targeting a different region of the genome, as was previously described.15 The sensitivity of wet mount in our study was found similar to that reported by others.34 This percentage is rather low for culture, although within the sensitivity range reported for this method, by other studies.35 In our study, PCR proved to be highly sensitive and specific, when compared to the other methods. Agreement between PCR and culture as well as wet mount, as expressed by the κ index was 0.79, which is considered satisfactory, although not the highest agreement index found between methods tested in this study.

In conclusion, our findings concerning the prevalence of trichomoniasis demonstrate a relatively low percentage in the female population living in Athens, a proportion of which have no symptoms or signs of vaginitis. The infection is more prevalent among the immigrant population. The study also supports the opinion that PCR, despite its cost and limited availability, could improve the diagnosis of T vaginalis, in comparison to traditional methods, whose sensitivity and specificity vary in several studies.

Take-home messages

  • The prevalence of trichomoniasis is low (4.6%) in the female population living in Athens, Greece.

  • The infection is more prevalent in symptomatic women (6.7%) than in asymptomatic ones (2.4%); it is also more frequently detected in immigrants (7.9%) than in Greek women (3.3%).

  • PCR improves considerably the diagnostic yield of T vaginalis, in comparison to traditional methods.



  • Funding This work was partially supported by the National University of Athens Special Account for Research Grants, Program “Kapodistrias” No. 70/4/4807.

  • Competing interests None declared.

  • Ethics approval This study was conducted with the approval of the Ethics Committee of Alexandra Hospital, University of Athens.

  • Patient consent Obtained.

  • Provenance and peer review Not commissioned; externally peer reviewed.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.