M ycoplasma genitalium is a sexually transmitted bacterium that causes nongonococcal urethritis in men and has been associated with cervicitis and pelvic inflammatory disease in women (1). The current recommended treatment of M. genitalium infection in Scandinavia is an extended course of oral macrolide. However, if empirical treatment of nongonococcal urethritis is initiated, usually only a single dose of azithromycin is given as treatment for suspected Chlamydia infection. Single-dose azithromycin treatment of M. genitalium infection is associated with a suboptimal treatment response and, in the case of treatment failure, with the development of M. genitalium macrolide resistance (2, 3). Macrolide-resistant M. genitalium strains are consequently prevalent, and resistance rates of 41% in a cohort of men with urethritis in London (4) and 38% in a recent population-based Danish survey (5) have been reported. In proven M. genitalium infection, macrolide susceptibility reporting is therefore necessary to guide therapy (6).Macrolide resistance in M. genitalium is strongly associated with the mutations A2058G and A2059G in the gene encoding 23S rRNA (3,7,8). In a recent Dutch study, a high proportion of the A2058T mutation causing resistance was observed (9). Resistance-associated mutations can be identified by PCR amplification and sequencing. This method is best suited to process samples in bulk and may result in unacceptable reporting times. Highresolution melting (HRM) analysis is another powerful tool to detect mutations in a target sequence, and PCR assays using HRM analysis for detecting mutations associated with macrolide resistance have recently been reported (10, 11). An alternative method for identifying known mutations is genotyping assays using differentially labeled hydrolysis probes targeting wild-type and mutated alleles (12,13). In the present study, we report such a 5= nuclease genotyping assay for the detection of the macrolide resistance-associated mutations in M. genitalium. The assay was used to determine macrolide resistance in 259 samples that had previously tested positive for the presence of M. genitalium, and the results obtained using this assay were compared to sequencing of the region of interest in the 23S rRNA gene. The 5= nuclease genotyping assay was highly accurate compared to sequencing.
MATERIALS AND METHODSM. genitalium detection. Between 8 October 2013 and 12 September 2014, the Department of Clinical Microbiology received 3,147 samples to test for the presence of M. genitalium. Testing was done using a laboratory-developed quantitative PCR (qPCR) using hydrolysis probes targeting the pdhD and mgpB genes of M. genitalium and an additional sampleprocessing control. The primers and probes of this laboratory-developed test have previously been published (14-16). In the multiplex qPCR, the final concentrations of M. genitalium-specific primers and of probes were 500 nM and 100 nM, respectively. Reactions were done in a total volume of 20 l with LightCycler 480 probes master (Roche A...