Our study, the first of M. genitalium in Singapore, revealed strikingly different prevalences of M. genitalium depending on the presence of Chlamydia trachomatis co-infection: 2.4% M. genitalium positive samples in CT negative individuals and 8.1% in CT positive patients. The DSC clinic is both the major government sexual health clinic and the site of the mandatory screening of licenced commercial sex workers (CSWs) in Singapore. It is a relative weakness of this study that we do not have precise demographic details for our cohort, but they will have comprised both CSWs and non-CSW patients undergoing CT/NG testing for clinical urethritis. As such, our data would be expected to over-estimate the prevalence of M. genitalium in the general population in Singapore. Previously reported prevalences of M. genitalium have varied considerably depending on the population studied and our findings compare with rates as low 1.2% in the general UK population [15] and as high as 22.4% in adolescent women in the US [16]. Within studies based on CSWs, or in sexual health clinics, reported rates of M. genitalium prevalence tend to be higher than those found in our study, leading to a conclusion of a relatively low prevalence of M. genitalium in Singapore, even in those at risk of infection [17]. This fits a general pattern of lower rates of Sexually Transmitted Infections (STI) in Singapore than the global average, however, it should be noted that this is in the context of an overall increase of STIs in the country [18].
Our study was designed to determine the relative prevalence of M. genitalium in CT/NG infected and uninfected patients. Such a correlation has been noted in some studies [15,16] but has been absent in others [19][14,15], but has been absent in other reports [20]. Our results show that, not only is there a heavy association of M. genitalium infection with CT infection in Singapore (odds ratio 3.5 as compared with CT PCR negative samples), but the odds ratio for this association (3.5) is much larger than previously published odds/hazard ratios of 1.9 – 2.5 [16,20,21]. While the association of CT and M. genitalium could result from shared risk factors for the transmission of these STIs, in our study, there was no association of M. genitalium with NG, which would also share similar risks for transmission, implying either a more a causal association between CT and M. gentialium infection, or at least a more nuanced epidemiological association.
The importance of considering CT co-infection while evaluating the burden of disease due to M. genitalium is highlighted by the concerning patterns of resistance in our samples. M. genitalium is very infrequently diagnosed in Singapore and thus very rarely specifically treated, whereas treatment of CT (or symptomatic urethritis) is common, with single-dose azithromycin. While single dose macrolide treatment was previously advocated for M. genitalium, evidence is growing to suggest that, in practise a five-day course of macrolides should be used to ensure M. genitalium eradication [7,22]. Resistance mutations in our study were exclusively found in M. genitalium isolates taken from patients with CT co-infection. Although involving relatively small datasets due to the low prevalence of M. genitalium in Singapore, our data strongly supports a conclusion that a single dose macrolide is indeed non-curative with respect to M. genitalium and that the treatment of CT is, at least to some extent, driving M. genitalium resistance. In this respect, our study could be improved with clinical data on our patients’ past exposure to macrolides to definitively show that repeated treatment of serially acquired or recurrent CT infections, or use of macrolides or fluroquinolones to treat other infections, is associated with M. genitalium antibiotic resistance in Singapore. Our data is unusual in that resistance to macrolides is usually much higher than that of fluoroquinolones [19,23–25], although predominance of fluoroquinolone resistance has previously also been noted [26,27]. With respect to the latter issue, we noted that although fluoroquinolones are rarely used as first-line treatment of NGU in the DSC clinic in Singapore, we still note a high resistance to these infections in our sample. Such resistance may be due to a high frequency of antibiotic use for other conditions, such as respiratory or urinary tract infection. A high general antibiotic pressure from fluoroquinolones in Singapore is implied by recent findings of identical resistance mutations in the 23S rRNA of M. pneumoniae [28] in Singaporean patients.
The high rate of overall resistance (37.75% of our samples resistant to either fluoroquinolones or macrolides) and multi-drug resistance (62.25% of our resistant samples had resistance to both macrolides and fluoroquinolones) follows the global trend of increasing levels of macrolide and fluoroquinolone resistance in M. genitalium [6,24,26,27]. The global situation of a slow, unnoticed accumulation of mutations to the only available therapies in patients with undiagnosed M. genitalium infection is concerning in light of the serious sequelae of untreated M. genitalium. The situation is compounded in Singapore, in which many of the newer antibiotics which have shown efficacy against M. genitalium are currently unavailable, in particular pristinamycin, a commonly recommended third line antibiotic [29]. Our results support the argument for close monitoring of antibiotic resistance in M. genitalium in Singapore, as in the rest of the world. Management of such resistance relies on timely diagnosis and appropriate treatment, ideally with subsequent confirmation of clearance of M. genitalium infections, but although diagnosis of M. genitalium has never been easier in light of novel and more convenient molecular tests [4], M. genitalium is often not part of routine screening or tested for in cases of NGU [8,30]. Evidently the ideal approach would be to consider M. genitalium in all cases of NGU, routine screening and monitoring of populations at risk of high M. genitalium transmission, and susceptibility-guided therapy of treatment failures [31]. However, where this is not feasible, we believe that the close association of M. genitalium and CT in terms of infection and antibiotic resistance suggests that ensuring adequate diagnosis and treatment of M. genitalium in CT infected patients may constitute a more cost-effective approach in a situation of finite resources. Furthermore, in light of the high rates of (multi-drug) resistance in the M. genitalium noted in this study we agree with recently published guidelines [32] that such treatment should extend beyond a single dose. The high frequency of dual -resistant strains of M. genitalium in our study highlights the urgent need to make some of novel antibiotics, such as pristinamycin, which are considered third-line options in other countries, available in Singapore. Overall, our data, revealing a low but significant prevalence of M. genitalium in Singapore, associated with a high level of resistance to the available treatment choices, highlight the importance of developing a robust approach to diagnosis and treatment of M. genitalium, whether general or targeted, to avoid the emergence of untreatable M. genitalium in the region.