Among men with genitourinary complaints, one of the most common conditions is urethritis, which is inflammation of the urethra. M genitalium infection is a frequent cause of non-gonococcal and non-chlamydial urethritis in men. In women, M genitalium infection is associated with endometritis, pelvic inflammatory disease, tubal factor infertility and possibly cervicitis (for review see Taylor-Robinson1 and references within). M genitalium is sexually transmitted and a cause of urethritis in men, which has been systematically shown by experimental inoculation of primates, by studies of concordance rates among partners, by antibiotic susceptibility studies and by studies of prevalence in men with and without urethritis.1^–7 M genitalium infection has also been associated with symptoms and signs of persistent non-gonococcal urethritis (NGU).4 8 Studies that report M genitalium in the urethra of men with NGU have either not included asymptomatic patients, not distinguished between patients who were symptomatic and those who were asymptomatic or have found an association with clinically symptomatic urethritis.4 5 7 9^–13 To our knowledge, there have been no reports of an association between M genitalium and NGU in men without symptoms of urethritis, probably because the study would require a large population sample. Studies of asymptomatic men with signs of urethritis is important because of recent recommendations and discussions for discontinuing routine urethral smears in asymptomatic men in genitourinary clinics.14^–19 The aim of this study was to determine the prevalence of M genitalium infection in clinically asymptomatic men with and without urethritis to understand the epidemiology of pathogenic M genitalium infection and to help determine risks associated with not screening asymptomatic patients.

MATERIALS AND METHODS

Study population and clinical methods

A retrospective analysis of 8468 men that had voluntarily attended a drop-in sexually transmitted infection (STI) clinic in Oslo without an appointment between November 2005 and December 2007, examined by a physician for urethritis, and tested for both M genitalium and C trachomatis was performed. Referral to a physician was made by a consulting triage nurse based on the following criteria: discharge, dysuria, or any other genital symptoms (for example, itch, rash, blisters or ulcers, warts), men who had sex with men (MSM), those who had many partners and inconsistent condom use, and men who were traced to a partner with a STI. For clinical diagnosis, a urethral smear was obtained with a sterile blunt curette, smeared on a glass slide, stained with methylene blue and examined microscopically. Patients were clinically diagnosed with urethritis for treatment if they had more than four polymorphic mononuclear leucocytes (PMNLs) per high-power field (HPF) in the absence of typical intracellular diplococci. For analysis in this study, patients were considered to have moderate to severe signs of NGU if they had more than nine PMNLs/HPF and borderline or negative for NGU if they had nine or less PMNLs/HPF. Men with diagnosed urethral gonorrhoea (n = 136) were excluded from analysis.

Laboratory methods

For detection of C trachomatis and M genitalium, DNA was isolated from 200 μl of first-void urine (FVU) using MagNA pure LC DNA isolation Kit I (Roche, Indianapolis, USA) and eluted into 100 μl of elution buffer. Altogether, 25 μl of the DNA preparation was used for detection of C trachomatis using Cobas Taqman 48 (Roche) and 10 μl of the same DNA preparation was used for detection of M genitalium by real-time PCR in a volume of 25 μl using primers and probes described by Jensen13 and using 7900 HT instrument (Applied Biosystems, Foster city, California, USA). Non-competitive primers and probes to an internal control template that was spiked into PCR master-mixes were used to assess inhibition in PCR reactions and to repeat tests if necessary.

Statistical analysis

Descriptive statistics were used to assess the characterisation of study participants, prevalence of microorganisms and clinical measurements using SPSS and Excel. When calculating prevalence of M genitalium (v.16) for association analysis, patients that were positive for C trachomatis were excluded from the analysis. Likewise, when calculating prevalence of C trachomatis for association analysis, patients that were positive for M genitalium were excluded. Measures of association by odds ratio (OR) between prevalence of microorganisms and patient characteristics, and clinical measurements and 95% confidence intervals (CIs), were performed using χ2 test.

RESULTS

In analysis of 8468 men that had been examined for urethritis, 989 patients (11.7%) were found to be positive for C trachomatis, 314 patients (3.7%) were positive for M genitalium and 44 (0.5%) were positive for both microorganisms. The latter co-infected group was excluded from further calculations. Altogether, 3024 (36%) patients had symptoms of urethritis (discharge and/or dysuria) and, of these, 1821 (60%) had moderate to severe microscopic signs of urethritis (>9 PMNLs/HPF). Out of 5400 patients without symptoms of urethritis, 2202 (41%) had moderate to severe microscopic signs of urethritis. Overall, the mean age was 31.3 (SD 8.9) years, which was similar to the mean age of patients positive for C trachomatis (28.4; SD 7.1), patients positive for M genitalium (31.8; SD 9.5) and patients with non-specific urethritis (NSU; 31.6; SD 9.1). Altogether, 7225 patients had reported having none to five sexual partners in the 6 months prior to their visit to the clinic. The mean number of partners for patients positive for M genitalium in this group was 2.2 (SD 1.3): slightly lower than for patients positive for C trachomatis (2.5; SD 1.3) and similar to patients with NSU (2.2; SD 1.3). Infection with M genitalium was associated with vaginal intercourse (OR 3.3; 95% CI 1.9 to 5.8) and unprotected sex in the last encounter (OR 1.6; 95% CI 1.1 to 2.4). M genitalium infection tended to be higher among men who had paid for sex (13.6%), although the number of patients who had documented payment for sex was too low for statistical significance or accuracy of patient information (n = 25; OR 3.1; 95% CI 0.9 to 10.5). MSM tended to have lower prevalence of M genitalium infection, as detected in FVU, compared with men who have sex with men and women (MSMW) or exclusively with women (MSW) (1.3% vs 3.2% vs 4.7%, respectively), although only the difference between the MSM and MSW was statistically significant (OR 3.7; 95% CI 2.1 to 6.4). A similar prevalence trend was also seen with C trachomatis among MSM, MSMW and MSW (6.1%, 6.7%, 13.4%, respectively). It should be noted that the average number of partners in the MSW group for the past 6 months was lower than MSMW or MSM (2.18 (SD 1.2), 3.13 (SD 1.3), 2.73 (SD 1.4), respectively) among patients who had reported having sex none to five times in the past 6 months, which adds significance to prevalence in the MSW group being the highest.

Prevalence of both organisms independently correlated with the severity of urethritis as quantified by microscopic signs of urethritis and irrespective of symptoms (fig 1). Both C trachomatis and M genitalium infection were also independently associated with symptoms of urethritis as diagnosed by dysuria and/or discharge (p<0.0001; table 1). Men were grouped into symptomatic and asymptomatic patients according to their anamnesis and analysed for association for moderate to severe microscopic signs of NGU (>9 PMNLs/HPF). Detection of both C trachomatis and M genitalium in urine was associated with moderate to severe microscopic signs of NGU in both symptomatic and asymptomatic patients (p<0.0001; table 2).

• Download figure
• Open in new tab
• Download powerpoint

Figure 1 Association of prevalence of Chlamydia trachomatis (A) and Mycoplasma genitalium (B) infection in patients with no (0–4 polymorphic mononuclear leucocytes/high-power field (PMNLs/HPF)), borderline (5–9 PMNLs/HPF), moderate (10–30 PMNLs/HPF) and severe (>30 PMNLs/HPF) microscopic signs of non-gonococcal urethritis. The number of patients positive for the microorganism out of total number of patients in each category is given along with the percentage, OR and 95% CI.

DISCUSSION

An association between infection with M genitalium and NGU has long been established in men since the application of the nucleic acid amplification test for detecting M genitalium.2 6 20 An association between clinical symptoms of NGU—that is, dysuria and discharge—and M genitalium infection has also been reported in men.5 12 Our data support previous findings that both M genitalium and C trachomatis are more frequently detected in symptomatic men than in asymptomatic men.5 12 21 22 To our knowledge, there have been no previous reports of an association between M genitalium infection and NGU, as measured objectively by quantification of PNMLs in urethral smears, in men without symptoms of NGU (see below for further discussions). A similar association has been shown for C trachomatis infection.23^–25 This finding is significant in arguing against a trend towards not taking urethral smears of men at STI clinics when they do not present clinical signs of NGU14 15 17–19 26–28 or to perform an alternative non-invasive test.29 In studies by Leung et al,12 an association between signs and symptoms of NGU and infection with M genitalium was reported, which is consistent with our findings in this study and other reported observations.5 12 21 22 30 However, in Leung et al’s 12 study, only eight patients with NGU and without discharge and/or dysuria had M genitalium infection and only three patients without NGU were infected with M genitalium—too few for association analysis in this category. Our analysis of a large number of men has allowed us to analyse a subset of asymptomatic men with microscopic signs of NGU adding further evidence to the causative role of M genitalium in NGU. However, it should be noted that the population described here is somewhat biased. While all people attending this STI clinic were consulted by a triage nurse and tested for C trachomatis, only patients that had any genital symptoms or those who were perceived to be at high risk of being exposed to STI were referred to examination by a doctor and tested for M genitalium. In a pilot study of 309 asymptomatic men seen only by the triage nurse, the prevalence of M genitalium and C trachomatis was 1.3% and 5.8%, respectively. In this geographical region, low prevalence of M. genitalium infection in men without risk factors or without urethritis does not support the cost-effectiveness of routine testing for M genitalium.

In this study, we found no association between the number of partners in the past 6 months and prevalence of M genitalium infection, although a pattern was present for patients positive for C trachomatis reporting none to five partners in the past 6 months. This is similar to reports by Falk,5 who also find no such association, and this may reflect a difference between the biology of C trachomatis and M genitalium infection. A longer patient anamnesis may be required to see such association. In this study, among 87 patients who had reported having no sexual partners in the last 6 months, the prevalence of M genitalium infection was 4.6%, similar to those reporting having one to five partners (4–4.8%). Prevalence of C trachomatis among patients who had claimed to have no sexual partners in the past 6 months was also 4.6% and it increased linearly to 16.8% among men who had reported having five partners in the past 6 months. C trachomatis has almost always been reported to be more frequent than M genitalium12 and, in this study, showing an association of C trachomatis with asymptomatic and symptomatic urethritis acts as an internal control in the general accuracy of clinical and laboratory findings and our interpretation of the data.

In summary, examination of urethral smears leads to identification of men who have the highest probability of being infected with M genitalium as well as C trachomatis. M genitalium infection is associated with NGU in asymptomatic as well as symptomatic men. Although routine testing of all young people attending STI clinics for C trachomatis is well-established in most west European countries, and negates the need for urethral smears in asymptomatic men, it seems likely that testing for M genitalium will be more difficult to establish as commercial and standardised tests for M genitalium are not available and lower prevalence rates of M genitalium may lead to routine testing of all patients to quickly fall out of favour. In our case, this would lead to not detecting 22% of all M genitalium infection cases. Taking urethral samples from men without clinical symptoms of NGU, but with other risk factors as described, will thus identify patients at high risk of infection with M genitalium and reduce the risk of spread to their female partners and, thereby, reduce the rates of pelvic inflammatory disease and infertility.