https://orcid.org/0000-0002-2557-8998
Ando et al. provide exciting data to support the use of prolonged sitafloxacin plus doxycycline for the treatment of highly resistant Mycoplasma genitalium infections.1 We should, however, be cautious in interpreting their reported cure rate of 78% because they do not stipulate whether these were symptomatic infections or not. They state that participants were recruited at the AIDS Clinical Center and the Sexual Health Clinic, Tokyo. Participants at the Sexual Health Clinic were all men who have sex with men (MSM) and were recruited while they were ‘regularly screened for HIV infection and other STIs’. This implies that the M. genitalium infections were predominantly asymptomatic. They do not state how participants at the AIDS Clinical Center were recruited. The ratio of rectal to urethral infections, however, suggests that most infections in the study were asymptomatic. Screening studies in MSM have found that most infections in this population are ano-rectal infections, the vast majority of which are asymptomatic.2,3 Ando et al. likewise, report that most of their infections were anorectal (78.8% and 81.4% of their two cohorts that received mono- and combination therapy, respectively). Several studies in MSM have found no association between the detection of M. genitalium in the anorectum and symptoms of proctitis.2,3
If the infections in their study were asymptomatic then this is relevant for two reasons. First, asymptomatic infections have a lower bacterial load and are therefore easier to clear with antimicrobial therapy.4–6 This means that clinical and microbiological cure rates may be lower in symptomatic infections.6 Second, if the results they report refer to infections detected by routinely screening asymptomatic individuals, then a reasonable case could be made that their intervention is causing the problem they claim to be solving. European M. genitalium clinical guidelines suggest limiting testing for M. genitalium to individuals with symptoms.7 This is because the equilibrium prevalence of M. genitalium in populations such as MSM attending STI clinics is ∼10% to 15%.7–9 This high prevalence is determined by the density of the sexual network in these population.9 As shown in Figure 1, this M. genitalium prevalence is unaffected by screening and treating. In this cohort of MSM taking HIV preexposure prophylaxis, screening the anorectum, pharynx and urethra of all participants every 3 months for 9 months had no impact on the prevalence of M. genitalium. Stopping screening after 9 months, however, resulted in a 38- and 2-fold decline in fluoroquinolone and macrolide consumption, respectively, and no appreciable increase in symptomatic or asymptomatic M. genitalium infections.8,9
M. genitalium prevalence is unaffected by screening. The prevalence of M. genitalium in this population of MSM taking HIV preexposure prophylaxis was stable at ∼10% during both the first 9 months of this study when individuals were screened and treated every 3 months and in the second 9-month period when the results of the M. genitalium screening were suppressed and only symptomatic infections were treated. Stopping screening resulted in a 38-fold reduction in fluoroquinolone consumption (based on9). This figure appears in colour in the online version of JAC and in black and white in the print version of JAC.
The major effect of screening and treating in this setting is thus to increase antimicrobial exposure, which is the major driver of antimicrobial resistance in M. genitalium and other bacterial species.10 These arguments have led us and others to suggest that the practice of screening for asymptomatic M. genitalium infection should be reconsidered in centres where this is still practised.7,8
Funding
This study was supported by internal funding
Transparency declarations
None to declare.
Author contributions
C.K. wrote the paper.
References
Author notes
Conflict of interests None declared.
