• mycoplasma genitalium
• pregnancy
• drug resistance, microbial
• molecular diagnostic techniques

Mycoplasma genitalium is a sexually transmitted bacterium that colonises the human urogenital tract and in pregnant women has been associated with adverse pregnancy and birth outcomes.1 In pregnancy, the prevalence of M. genitalium is estimated to be 0.9% (95% CI 0.6% to 1.4%) in high-income settings and 12.5% in individual studies in low- and middle-income-countries.1 2 Globally, M. genitalium antimicrobial resistance (AMR) continues to escalate,3 and despite rising rates of AMR the prevalence of macrolide and fluoroquinolone resistance in M. genitalium has not been widely studied among pregnant women.2 Treatment of M. genitalium during pregnancy with selected antimicrobials such as macrolides is recommended, while tetracyclines are contraindicated and caution is recommended for fluoroquinolones,4 highlighting the potential issues surrounding available/effective antibiotics for this pathogen during pregnancy. In Papua New Guinea (PNG), macrolides and fluoroquinolones are routinely used for treatment of STIs4 and hence we aimed to characterise M. genitalium AMR in pregnant women.

We screened 69 M. genitalium-positive urine samples obtained from women attending their first antenatal clinic visit at five health facilities in Madang Province (PNG) between 2018 and 2019, enrolled in the control arm of a randomised controlled trial in PNG, the Women And Newborn Trial of Antenatal Interventions and Management.5 Confirmatory PCR was performed to verify the presence of M. genitalium in all 69 samples. Samples with positive results were then tested for the presence of AMR using targeted PCR assays. Macrolide and fluoroquinolone resistance/susceptibility markers were determined using previously published assays (online supplemental file). Stata V.15.1 was used to calculate the proportion of samples with resistance markers, with 95% exact binomial CI calculated using the Clopper-Pearson method. Where there were zero events, we give the 97.5% CI as the upper limit.

From the five study sites, all 69 samples that were initially reported to be positive for M. genitalium were confirmed as positive by repeat PCR testing. Of these, 58 (84.1%) samples were successfully characterised for macrolide resistance, while 54 (78.3%) were successfully characterised for fluoroquinolone susceptibility/resistance (table 1). In summary, no samples with available results harboured macrolide resistance mutations (0.0%, 975% CI 0.0% to 6.2%) or the S83I fluoroquinolone resistance mutation (0.0%, 97.5% CI 0.0% to 6.6%), and all 54 samples were characterised as fluoroquinolone-susceptible (S83 wild-type).

While data on M. genitalium AMR remain limited in many regions, including in low-income countries of the Western Pacific region, our findings indicate no evidence of macrolide or fluoroquinolone resistance in PNG. This further supports the findings of two previous studies of M. genitalium in PNG (78 pregnant women) and Solomon Islands (56 non-pregnant women) that also reported no macrolide resistance.2 6 Some of the reasons behind the lack of resistance in M. genitalium in these regions may include the fact that current STI treatment guidelines in PNG do not stipulate testing and treatment of M. genitalium in symptomatic persons.4 Other studies have identified that incorrect screening practices leading to treatment with macrolides have contributed to M. genitalium’s increasing resistance to macrolides.7 Given that both macrolides and fluoroquinolones are prescription drugs in PNG, their use in treating STIs and other infections is highly regulated. These drugs are not available for use in self-medication and so the repeated consumption/use of these drugs is not likely to be driving the selection of resistance in PNG, as has been suggested in other regions of the world.

From an STI management perspective, the clinical symptoms of M. genitalium infection are not distinct nor fully understood, as M. genitalium can be found in conjunction with other STIs, including gonorrhoea and chlamydia. In PNG, symptomatic patients often present with more than one curable STI. In this setting, the targeted testing and treatment for specific STIs is not routine and instead syndromic management of STIs is the standard of care. Resource-constrained health facilities throughout PNG use a single-dose cocktail of antimicrobials, including augmentin, amoxicillin, probenecid, tinidazole and azithromycin/erythromycin. An increase in presumptive treatment of these infections could result in the overuse of macrolides, which could indirectly lead to the development of AMR in M. genitalium. Additionally, asymptomatic women in PNG are less likely to attend sexual health clinics and therefore unlikely to receive either macrolides or fluoroquinolones, reducing the potential for de novo AMR.8

It is important to report on findings where AMR has not been detected to reduce the risk of overestimation of M. genitalium resistance in these settings. Future studies of M. genitalium AMR in PNG and similar settings are warranted, given the potential for import of resistant infections or for de novo resistance, if the use of macrolides and fluoroquinolones for reproductive tract infections increases.