Article Text


Epidemiology oral session 3: bacterial resistance
O1-S03.06 Evidence of circulating macrolide resistance in Mycoplasma genitalium infections and development of a rapid assay to detect resistance
  1. J Twin1,
  2. C Bradshaw2,
  3. S Garland3,
  4. C Fairley2,
  5. J Jensen4,
  6. S Tabrizi3
  1. 1Murdoch Children's Research Institute, Melbourne, Australia
  2. 2Melbourne Sexual Health Centre, Australia
  3. 3Royal Women's Hospital, Australia
  4. 4Statens Serum Institut, Denmark


Background Mycoplasma genitalium (Mg) is a sexually transmitted bacterium causing urethritis, cervicitis and longer term sequelae such as pelvic inflammatory disease. At Melbourne Sexual Health Centre (MSHC), Australia, from December 2007 to December 2009, 111 M genitalium infected patients (86 males/25 females) were treated with 1 g azithromycin with a cure rate of 69% (95% CI 60% to 77%). Resistance to macrolide antibiotics such as azithromycin occurs via mutations of the Mg 23S rRNA gene in response to exposure of sub-therapeutic doses of the drug. It is our hypothesis that the unacceptably high rate of treatment failures seen at this clinic is due to macrolide resistant Mg strains circulating in the community. To test this theory, we determined whether resistance was present in initial consult samples collected at MSHC, and we developed a rapid assay able to detect resistance during routine diagnostics.

Methods A subset of 83 Mg positive samples from patients taken prior to azithromycin was analysed in this study (62 males/20 females); 56 of these patients were subsequently cured by azithromycin and 27 cases failed azithromycin. DNA sequencing was then carried out to determine each respective 23S rRNA sequence type. A real-time PCR assay coupled with high resolution melt analysis was developed to detect the mutational changes in the Mg 23S rRNA gene, dubbed MARS (Macrolide Antibiotic Resistance Screen), and was tested against a panel of known macrolide resistant Mg isolates and the subset of samples from MSHC.

Results In total, 16/83 (19%) of the pre-treatment samples tested possessed 23S rRNA mutations conveying macrolide resistance; significantly more patients who failed 1 g azithromycin had pre-existing macrolide mutations (12/27; 44%) compared to those who were cured by 1 g azithromycin (4/56; 7%); p<0.0001 (Abstract O1-S03.06 table 1). The MARS assay that was developed was able to identify when a 23S rRNA mutation was present in 100% of these samples.

Abstract O1-S03.06 Table 1

23S rRNA gene mutations present in a subset of 2007–2009 Mg positive samples from initial consult from MSHC

Conclusions This data shows compelling evidence that macrolide resistance is circulating in certain populations and is attributing to a significant level of treatment failures seen in an Australian sexual health clinic. The development of a rapid molecular assay to detect resistance provides the means for clinicians to choose a more appropriate second line treatment option such as moxifloxacin, and thereby reduce transmission of resistant strains and avoiding sequelae associated with persistent infection.

Statistics from

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.