Article Text


Unusually low prevalence of Mycoplasma genitalium in urine samples from infertile men and healthy controls: a prevalence study
  1. Vanda Plecko1,
  2. Lidija Zele-Starcevic1,
  3. Vesna Tripkovic1,
  4. Mihael Skerlev2,
  5. Suzana Ljubojevic2,
  6. Sanja Plesko1,
  7. Ivana Marekovic1,
  8. Jorgen Skov Jensen3
  1. 1Department of Clinical and Molecular Microbiology, Clinical Hospital Centre Zagreb and School of Medicine University of Zagreb, Zagreb, Croatia
  2. 2Department of Dermatology and Venerology, Clinical Hospital Centre Zagreb and School of Medicine University of Zagreb, Zagreb, Croatia
  3. 3Mycoplasma Laboratory, Staten's Serum Institute, Copenhagen, Denmark
  1. Correspondence to Dr Vesna Tripkovic; vbtripkovic{at}


Objective To detect Mycoplasma genitalium in urine samples of infertile men and men without any signs of infection in order to investigate whether M. genitalium and other genital mycoplasmas (Mycoplasma hominis and Ureaplasma spp) are found more often in urine samples of infertile men than in asymptomatic controls and to determine resistance to macrolides.

Methods The study included first void urine samples taken from 145 infertile men and 49 men with no symptoms of urethritis. M. genitalium, Chlamydia trachomatis and Neisseria gonorrhoeae were detected by commercial PCR. Trichomonas vaginalis was detected by microscopy and culture. M. hominis and Ureaplasma spp were detected by culture. M. genitalium was detected by in-house conventional and real-time PCR.

Results Two M. genitalium positive samples were found among samples obtained from infertile men. All asymptomatic men were M. genitalium negative. Macrolide resistance was not found in either of the two positive samples.

Conclusions In comparison with reported data, an unusually low prevalence of M. genitalium was found in infertile men. The reasons for this unexpected result are not known; possibly, local demographic and social characteristics of the population influenced the result. Further studies to investigate M. genitalium in infertile and other groups of patients are needed.

Statistics from

Strengths and limitations of this study

  • We are aware of low number of urine samples in both groups of participants.

  • These are only preliminary results, and were unexpected in comparison with reported data. However, the results were confirmed in two independent laboratories to exclude the possibility of laboratory error.

  • We plan to continue this study, collecting urine samples from a larger number of infertile men and comparing them both with men with urethritis and asymptomatic men.

  • If the results are similar (ie, an unexpectedly small number of men with Mycoplasma genitalium), we will try to determine the reason for this.


Reliable detection of Mycoplasma genitalium became possible after the development of PCR assays.1 ,2 The prevalence of M. genitalium in patients with non-gonococcal urethritis (NGU) ranges from 13% to 42%; and in asymptomatic men from 0% to 15%.24 The impact of M. genitalium on male fertility remains unclear.2 ,5

Our aim was to detect M. genitalium, M. hominis and Ureaplasma spp in first void urine (FVU) samples of infertile men and men without any symptoms and/or signs of infection and, additionally, to determine the prevalence of macrolide resistance in M. genitalium. We restricted our study to infertile men without the following common sexually transmitted infections (STIs): Chlamydia trachomatis, Trichomonas vaginalis and Neisseria gonorrhoeae.

To our knowledge, this is the first study in Croatia which has been undertaken to detect M. genitalium.


The study was approved by the ethics committee of the School of Medicine, University of Zagreb. It is part of the Croatian Ministry of Science grant (108-1080114-0014): “Molecular detection of microorganisms: their influence on antimicrobial consumption”.

Each participant provided written informed consent, and completed a questionnaire stating reasons for attendance, age, symptoms of urethritis, number of lifetime sexual partners, history of STIs and recent/current antibiotic treatment.

FVU (about 20 mL) was taken from the patients; 4–5 mL of each sample was used for culture of M. hominis, ureaplasmas and T. vaginalis. For PCR detection of M. genitalium, C. trachomatis and N. gonorrhoeae, 4–5 mL of each sample were used. Five millilitres of the original FVU were frozen and shipped to Staten Serum Institut in Copenhagen for confirmation by real-time M. genitalium PCR. Samples were immediately processed for M. hominis, T. vaginalis and ureaplasmas. For PCR detection samples were stored at −20°C.

C. trachomatis and N. gonorrhoeae were detected by PCR (Cobas TaqMan CT/NG Test, V.2.0 Roche Diagnostic, Basel, Switzerland), as described by the manufacturer, and urethral swabs were obtained for detection of N. gonorrhoeae by culture (BBL MTM, New Jersey, USA). T. vaginalis was detected by microscopy and culture in modified Diamonds medium (Remel, Inc, Santa Fe, USA). Thirteen of the infertile men and three controls were excluded owing to infection with a recognised STI (Chlamydia trachomatis, Trichomonas vaginalis and Neisseria gonorrhoeae), leaving a total of 194 FVU samples. These were collected in polypropylene containers (Sarstedt, Nümbrecht, Germany) from men who were referred to the Department of Clinical Microbiology and Department of Dermatology, Clinical Hospital Centre Zagreb. One hundred and forty-five samples were obtained from men as a part of an infertility investigation. None of them had any symptoms of genitourinary infections, but all had an abnormal semen quality analysis (oligozoospermia, asthenozoospermia, oligoasthenozoospermia). Infertility of their female partners was excluded by hysterosonosalpingography.

Forty-nine samples were from asymptomatic men attending the clinic as a part of an annual physical examination.

Aliquots of the urine samples (4–5 mL) were prepared for culture of genital mycoplasmas and for molecular testing, respectively. Urine samples were centrifuged at 3000×g for 5 min and sediments were inoculated in urea–arginine broth (bioMerieux, Lyon, France) and onto A7 agar (Becton Dickinson, Cockeysville, Maryland, USA). The vials were incubated for 48 h at 37°C. The agar plates were incubated at 37°C in 5% CO2 for 5 days and examined microscopically for the appearance of typical mycoplasma colonies.6

The rest of the urine (4–5 mL) was concentrated at 20000×g for 15 min at 4°C. The pellet was resuspended in 200 µL of 20% w/v Chelex 100 slurry (Sigma, USA) in TE buffer (10 mM Tris-HCl, pH 8.0, 1 mM EDTA). The mixture was vortexed for 1 min; then placed in a thermoblock for 10 min at 95°C. The mixture was centrifuged briefly and the supernatant aspirated into new tubes and stored at −20°C until required for PCR.7

PCR for detection of M. genitalium was performed with two pairs of primers: the first targeted the 16S rRNA gene: 16SFG2 (5′-CCT TAT CGT TAG TTA CAT TGT TTA A), 16SRG (5′-TGA CAT GCG CTT CCA ATA AA),7 and the second targeted the MgPa major adhesin gene: MgPa1 (5′-AGT TGA TGA AAC CTT AAC CCC TTG G), MgPa3 (5′-CCG TTG AGG GGT TTT CCA TTT TTG C).1

All samples were examined by both assays with internal controls for PCR inhibition, and both PCRs were performed as previously described.1 ,79

The PCR was performed in an automated DNA thermal cycler (PCR System 9700, Applied Biosystems).

To confirm the results, an aliquot of the original FVU samples (5 mL) was shipped to Statens Serum Institut, Copenhagen, Denmark, where it was tested by an inhibitor-controlled real-time PCR using primers detecting the MgPa gene, as previously described.7 ,9

Macrolide resistance mediating mutations in region V of the 23S rRNA gene was detected by DNA sequencing of amplicons obtained directly from the clinical specimens, and performed at Statens Serum Institute Copenhagen, Denmark.10

STATISTICA (data analysis software system), V.10 (StatSoft, Inc (2011), USA) was used for data analysis. The median was used to describe the age of groups and the number of observations and percentage to describe categorical variables. To compare age between the groups a Mann–Whitney U test was used, and to compare categorical variables a χ2 test was used. A p value of <0.05 was considered to be statistically significant for all tests performed.


The infertile men were comparable to the controls for age (z=−0.805, p=0.421, Mann–Whitney U test). They significantly more often reported a history of STIs (χ2=14.443, df=1, p=0.0001) and a higher number of lifetime sexual partners (χ2=35.734, df=2, p<0.0001; table 1).

Table 1

Demographic and epidemiological data for all groups

Thirteen of the infertile men and three controls were excluded owing to infection with a recognised STI (Chlamydia trachomatis, Trichomonas vaginalis and Neisseria gonorrhoeae).

Among the infertile men one patient had N. gonorrhoeae infection, one patient was diagnosed with T. vaginalis and 11 patients were C. trachomatis positive. These samples were also tested for M. genitalium, but all were negative.

Ureaplasma spp and M. hominis were isolated from the same proportion of infertile men and asymptomatic controls (χ2=0.435, p=0.509; χ2=0.021, p=0.886), respectively. Only two samples were positive for M. genitalium; both in the group of infertile men (2/145; 1.4%; 95% CI 0.2% to 4.9%). These men were aged 29 and 37 years, respectively, and reported three and seven lifetime sexual partners compared with the majority of the group who had had 5–10 partners.

In our laboratory we used conventional in-house PCR (qualitative) and results were confirmed at the Staten Serum Institut in Copenhagen, Denmark by real-time PCR. All M. genitalium results were concordant when the samples were examined by real-time PCR in Copenhagen. M. genitalium load for two positive samples was 778 copies/mL (c/mL) and 6765 c/mL, respectively. The man with a M. genitalium load of 778 c/mL was diagnosed with oligozoospermia, and the other (M. genitalium load of 6765 c/mL) with asthenozoospermia.

Ureaplasma spp were found in 30% (43/145) of the infertile men compared with 35% (17/49) of the asymptomatic men. M. hominis was positive in 21% (31/145) and 20% (10/49) of infertile men and asymptomatic men, respectively.

In 12 samples from the infertile men co-infection with Ureaplasma spp. and M. hominis was found. In the group of asymptomatic men co-infection in three samples was found. In the two samples with positive M. genitalium, taken from infertile men, no other pathogens were present.

Macrolide resistance mediating mutations in the 23S rRNA gene of M. genitalium were not found in either of the two positive samples.


This study demonstrates a low prevalence of M. genitalium in infertile men in the Zagreb region, Croatia. Ureaplasmas and M. hominis were often detected in both infertile men and healthy controls, suggesting that they should be considered commensals.

FVU samples were used because several studies have reported that molecular methods performed on urine samples can detect as many, or even more, infected patients than traditional urethral swabs, or cervical swabs or semen.5 ,7 No data for the prevalence of genital mycoplasmas in Croatia exist.

In this study the prevalence of Ureaplasma spp did not differ significantly among infertile men and asymptomatic controls, and was present in about one-third of both groups. This strongly suggests that ureaplasmas do not have a significant role of in male infertility.

We did not perform a specific test for the Ureaplasma spp. Most of the published studies have reported the prevalence of ureaplasmas in infertile men without discriminating between U. urealyticum and U. parvum. The data are not conclusive about the prevalence of U. urealyticum and U. parvum. Abusaraha et al,11 found that U. parvum was the most prevalent isolate detected among infertile men (90%).

M. hominis was detected in 20% of asymptomatic men and 21% of infertile men, respectively, a higher prevalence than in some other studies.12 M. hominis is considered normal flora of the urethra and the prevalence of M. hominis may reflect a high prevalence of bacterial vaginosis in the men's sexual partners, as M. hominis is known to be strongly associated with this condition in women.13

The prevalence of M. genitalium varies significantly in different populations2 and was low in our study. Other studies have also found that M. genitalium is uncommon in the FVU of infertile men.5 The two positive samples in our study were from the group of 145 infertile men and M. genitalium was not detected in any of the controls. We were concerned that technical problems might have caused the low prevalence, and therefore, frozen FVU samples were shipped to Copenhagen for evaluation. However, a 100% concordance between the results was found, suggesting that the prevalence of M. genitalium in this Croatian population is, indeed, very low.

A possible relationship between infection and infertility has been the subject of controversy for years. It is estimated that only 15% of male infertility is related to genital tract infection.5 ,14 Detection of bacteria in urogenital samples does not necessarily suggest infection but may signify colonisation, contamination or infection.

Only a few studies have examined the association between M. genitalium and male infertility5 ,15 and these studies did not have control populations of fertile men. We tried to design a study in which all other potential infective causes of infertility were excluded. In recent studies the prevalence of M. genitalium in infertile men was similar to the prevalence found by us (1.4%): 4.8% in the study of Gdoura et al5 and 3.2% was in the study of Abusaraha et al.11

Findings for C. trachomatis, which is the most common bacterial cause of NGU, were similar. C. trachomatis in women is a well-established cause of tubal factor infertility; in men it causes NGU. Also, it has been shown that C. trachomatis attaches to spermatozoa (on the surface and in the nucleolus). However, its role in male infertility, like the role of M. genitalium, is not yet clear. There are significant variations in the prevalence of C. trachomatis infections in men with infertility ranging from 0% to 42.3%, depending on the methodology, type of sample and differences of infection rates in different populations. In a recently published Canadian study16 the prevalence of C. trachomatis infection in 5588 infertile men, was 0.3%. The author concluded that this low prevalence clearly demonstrates that a small proportion of male infertility is caused by C. trachomatis.17

We attempted to study an asymptomatic group of men without urethritis as controls, and found that all were negative for M. genitalium. Unfortunately, the infertile men had had significantly more partners and also reported previous STIs more commonly than did the controls, suggesting that the control group had less risky behaviour.

Both M. genitalium positive samples were tested for macrolide resistance and were susceptible. This is encouraging considering the widespread use of azithromycin in the treatment of chlamydia and unspecific urethritis in Croatia. It is not possible to provide estimates of the prevalence of resistance to macrolides in this bacterium. Obviously, more M. genitalium positive samples should be tested in order to guide future treatment guidelines.

An unusually low percentage of M. genitalium was found in this study. The reasons for this unexpected result cannot be explained. Further studies to investigate M. genitalium in infertile and other group of patients from Croatia are needed.


We thank Birthe Dohn for excellent technical assistance with M. genitalium testing at Statens Serum Institut, Copenhagen, Denmark.


View Abstract


  • Contributors VP designed the study, was responsible for questionnaire data and data analysis, and wrote part of the manuscript; LZ-S tested samples for Mycoplasma genitalium, and drafted the article; VT planned the study and wrote part of the text; MS collected samples and edited the manuscript; SL collected samples and questionnaire data; JSJ tested samples for M. genitalium, edited the manuscript and approved the final version; SP tested samples and corrected the draft version; IM revised the manuscript.

  • Funding The study is part of the Croatian Ministry of Science grant (108-1080114-0014): “Molecular detection of microorganisms: their influence on antimicrobial consumption”.

  • Competing interests None.

  • Patient consent Obtained.

  • Ethics approval Ethics committee, School of Medicine, University of Zagreb.

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

  • Data sharing statement No additional data are available.

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.