In our population of adult men and women attending an STI clinic,the prevalence of MG was 7.2% (7.4% men, 6.9% women), and 38.8% of MG infected patients were symptomatic. Our study from an STI clinic in GuangDong Province adds to the sparse data of MG prevalence in China. Our prevalence rate is higher than those found in patients attending STI clinics with symptoms of urogenital tract infections or for a check up in Greece (5.7%) [29] and Norway (4.9%) [30], but lower than that of Denmark (9.0%) [30], Sweden (9.8%)[30] and US (16.7%)[31].
In the investigation of MG in men attending an STI clinic with or without symptoms in other countries, the prevalence were still low in Norway (4.5%)[30], similar with our result showed 8.9%, 9.1% and 6.4% in Denmark [30], Sweden [30] and Greece [29]. However, those were much higher in Russian (17%) [32] and US (17.2%) [31]. For men with urogenital symptoms, the MG positive rates were 12.8%(17/133), 12.3% and 16.7% in our study, London[33] and Brussels [34], respectively. In MSM, the prevalence of MG was 9.3% in three anatomical sites (genital/pharyngeal/rectal) in London[33] and 5.4% in the rectal samples in San Francisco[35].
The MG prevalence of women in our study was 6.9%(95%CI:3.2%-10.5%), which was similar with those found in women with symptoms of urogenital tract infections or for a check up in Greece (6.9%) [29] and Norway (6.0%)[30], but lower than that of Denmark (9.3%) [30], Sweden (11.1%)[30] and two researches in US (16.3%, 17.5%)[31, 36]. Among female sex workers, the MG prevalence rate was 13.2%, 12.7% and 22.4% in China[27], Kenya[37] and Honduras [38], respectively. For females at low risk of acquiring STI, the prevalence of MG was low: 1.5% in Louisiana[39] and 2.3% in London[40].
In our study, symptomatic patients were more likely to have MG infection compared to asymptomatic individuals(12.5% Vs 5.7% ,p = 0.004). This was also reported in men but not in women for patients attending a young person clinic in Sweden[41]. However,some studies had the opposite conclusions, such as no statistical association between clinical symptoms and MG infection in men from an STI clinic in Guangxi Province in China[26], and no association between MG infection and microscopically defined urethritis or cervicitis in Greece[29]. These different conclusions may be related to the composition of the sample and the definition of symptoms.
In studies from Greece[29], Kenya[37] and Honduras[38], consistent condom use was not protective against MG. However, it significantly decreased the association of MG infection in our study (p = 0.043). Patients who reported not using a condom all the time in the past 6 months were more likely to be infected with MG than patients who reported using condom all the time in the past 6 months. Thus, consistent condom use is still highly recommended among sexually active individuals, especially in the high-risk population.
In our study, no association with other STIs was found in MG infection using the Chi-square test analysis, consistent with previous findings from UK[33, 42] and Greece[33, 42], showing that MG might be an independent pathogen in the genital tract. The high level of MG and CT/NG co-infection suggests that screening and treating CT may not have too much impact on MG since azithromycin 1 g which is the first-line drug in treating CT appears to be a suboptimal choice in MG treatment. In MG treatment, antimicrobial resistance (especially for macrolide resistance) has been reported[43, 44]. To avoid exacerbating more antimicrobial resistance, many resistance-guided therapy for MG infections was more effective[45]. In resource limited settings, where resistance could not be routinely tested, modifying the existing treatment guidelines and making the treatment more targeted are urgently needed.
One of the limitations of our study include the inability to recruit larger numbers of female patients and small numbers of individuals with MG, the other limitation is that we recruited more patients with genital warts as our clinic has a large volume of patients with genital warts, and this could have weakened the analysis of association for MG infection. The results reflect the epidemiology of MG in Guangdong for individuals attending an STI clinic but may not be generalized to all patient populations in China.
In conclusion, MG appears to be a relatively common infection among individuals attending an STI clinic in Guangdong(7.2%). Symptomatic patients were statistically significantly associated with MG detection and consistent condom use was negatively associated with MG detection. MG did not appear to be associated with other STIs according to our analysis. Further research is needed to better understand its global epidemiology and natural history.