LAUSR

Macrolide resistance in Mycoplasma genitalium infection is emerging worldwide and is largely driven by use of azithromycin in STI treatment. South Africa has used azithromycin in its syndromic management regimen of male urethritis and vaginal discharge since 2015, but prevalence of macrolide resistance in M. genitalium remains largely unknown. This study determined azithromycin resistance in M. genitalium in remnant vulvovaginal specimens that had been obtained from pregnant women in Cape Town, South Africa, between November 2017 and February 2019. In brief, vulvovaginal swabs were selfcollected at participants’ first antenatal care (ANC), third trimester ANC and postnatal care visits. Infacility GeneXpert testing (Cepheid, Sunnyvale, California) for Chlamydia trachomatis, Neisseria gonorrhoeae and Trichomonas vaginalis was done, followed by treatment if indicated. The Aptima Vaginal Swab Specimen Collection Kit (Hologic, San Diego, California) was used to collect a second swab that was stored at −20°C at the National Institute for Communicable Diseases for batched testing of M. genitalium using the Aptima M. genitalium Assay (Hologic) after study completion. All specimens with a positive Aptima assay result were evaluated for presence of macrolide resistance mutations using the ResistancePlus M. genitalium assay (SpeeDx, Australia) as per manufacturer’s instructions at the University of Pretoria. M. genitalium was detected with the Aptima Assay in 84 specimens from 38 women in the study cohort. The median age of these women at baseline was 28 years (range 19–40 years); 24 (63%) were HIV infected and 19 (50%) had another STI: C. trachomatis (n=9), N. gonorrhoeae (n=2) and T. vaginalis (n=10). Resistance testing with the ResistancePlus assay was successful in 64/84 (76%) specimens; these had been obtained from 34/38 women (89%) with M. genitalium infection detected at either first ANC (26/35, 74%), thirdtrimester ANC (22/26, 85%) and postnatal visit (16/23, 70%). Wildtype M. genitalium was detected in all 64 (100%) specimens. Specimen degradation likely influenced the recovery rate for resistance testing in our study because M. genitalium is generally a lowload infection, and macrolide resistance testing was done after 14 months of specimen storage (median, IQR 11–16 months) and after 6 months and at half the input volume of the Aptima assay. Differences in assay sensitivity might also play a role. To our knowledge, M. genitalium bacterial load and macrolide resistance are not associated, so we consider selection bias unlikely. This study confirms that macrolide resistance is uncommon in M. genitalium infections in South Africa despite the use of azithromycin for syndromic STI management since 2015. It is the first data point from the Southern part of the country, >1000 km away from the settings of previous studies (table 1), suggesting that geographical differences in resistance are not present in the country. Nevertheless, continuous surveillance is warranted as emergence of resistant M. genitalium could undermine the effectiveness of syndromic STI management.