Article Text
PDF
Abstract
Objectives Mycoplasma genitalium is an important cause of STI in men and women. Worldwide evidence suggests a reduction in efficacy of azithromycin treatment due to the prevalence of macrolide resistant M. genitalium. The aim of this study was to describe the prevalence of macrolide resistance in patients with a positive test for M. genitalium within our setting.
Methods Two STI clinics in Stockholm offered tests for M. genitalium as part of a routine care pathway. Positive specimens were analysed for macrolide resistance mediating mutations by sequencing.
Results During the study period, 171 (7.5%) of 2276 patients had a positive M. genitalium test; 7% of women and 8% of men. Macrolide resistance was detected in 31 (18%) of the M. genitalium positive; treatment with azithromycin within the previous 6 months was strongly associated with macrolide resistance.
Conclusions The prevalence of macrolide resistance was lower in Sweden than in other Northern European settings. We hypothesise that this observation may be due to use of doxycycline as primary treatment of Chlamydia trachomatis. The efficacy of empirical treatment is challenged by azithromycin resistant M. genitalium. Clinically available and enhanced diagnostics targeting this pathogen are urgently required. We suggest a test of cure 3–4 weeks after start of azithromycin therapy since macrolide resistance develop during treatment.
- M GENITALIUM
- ANTIMICROBIAL RESISTANCE
- SURVEILLANCE
- GENITAL TRACT INFECT
Statistics from Altmetric.com
Introduction
Several studies have demonstrated a worldwide emerging problem with increasing macrolide resistance in Mycoplasma genitalium.1 The resistance mediating mutations have been located in region V of the 23S ribosomal RNA.2 In Scandinavia, increasing problems with treatment failure after azithromycin are experienced. A retrospective case study in Sweden reported that macrolide resistance before treatment had increased from 0% in 2006 to 18% in 2011,3 and a nationwide retrospective survey in Denmark detected macrolide resistance in 38% of unselected patients.4
Methods
This study was approved by The Regional Ethics Review Board at Karolinska Institutet in Stockholm (approval number 2011/1259-32). Samples were collected throughout 2012 in two STI clinics in Stockholm (at the Karolinska University Hospital in Huddinge and at the South Hospital in Stockholm). Patients attending for STI testing were offered tests for M. genitalium independently of the reason for attending as part of routine patient management. In men, first void urine specimens (approximately 10 mL of a 30 mL portion) and in women first void urine specimens (approximately 10 mL of a 30 mL portion) combined with swabs from the vagina and/or cervix were used. The specimens were analysed at the Karolinska University Laboratory with a PCR method.5 Positive specimens were sent to Statens Serum Insitut in Copenhagen for detection of macrolide resistance mediating mutations using a pyrosequencing assay.4
Differences in proportions were compared using Fisher's exact test and 95% CIs were calculated for percentages.
Results
A total of 2276 patients were screened in this study, 980 women and 1296 men. Overall, 171 patients (7.5%; 95% CI 6.5% to 8.7%) had a positive test for M. genitalium and in 31 patients (18%; 95% CI 13% to 25%) macrolide resistance was detected. Resistance mediating mutations were found in 17 of 72 samples (24%; 95% CI 14% to 35%) from women and in 14 of 99 (14%; 95% CI 8% to 23%) of the men (p=0.16).
From the South Hospital 1741 patients were included, 795 women and 946 men and 6% (95% CI 5% to 7%) were positive for M. genitalium (49 women and 56 men); macrolide resistance was detected in 24% (95% CI 13% to 39%) of the women and in 9% (95% CI 3% to 20%) of the men. The Karolinska University Hospital included 535 patients, 185 women and 350 men and 12% (95% CI 10% to 15%) were positive for M. genitalium (23 women and 43 men); macrolide resistance was detected in 22% (95% CI 7% to 44%) of the women and in 12% (95% CI 4% to 25%) of the men. Of the 171 M. genitalium positive patients, 150 (88%; 95% CI 82% to 92%) were seen by a doctor, mostly due to symptoms and 21 (12%; 95% CI 8% to 18%) by a midwife or a nurse. 47% (95% CI 44% to 50%) of the women and 64% (95% CI 61% to 67%) of the men were symptomatic at attendance.
Information regarding a self-reported previous M. genitalium infection was available for 155 of the 171 M. genitalium positive patients (30 with macrolide resistance and 125 without) (table 1). A significantly higher proportion of patients with a previously diagnosed M. genitalium infection were infected with strains carrying macrolide resistance mediating mutations (23 of 40 (58%; 95% CI 41% to 72%) compared with 7 of 115 (6%; 95% CI 2% to 12%)) p<0.0001. Similarly, information regarding self-reported azithromycin within the last 6 months was available for 87 patients (23 with macrolide resistance and 64 without) (table 1). Recent treatment with azithromycin was strongly associated with macrolide resistance as 20 of 22 (91%; 95% CI 71% to 97%) of patients having received azithromycin were resistant compared with 3 of 62 (5%; 95% CI 1% to 13%), p<0.0001 of the azithromycin naïve.
Relation between previous Mycoplasma genitalium infection and azithromycin use among 171 infected patients distributed according to resistance status
Discussion
Several studies worldwide have demonstrated that M. genitalium causes STI as urethritis, cervicitis and pelvic inflammatory disease6 and others have demonstrated the growing problems with development of macrolide resistance.1 Our results confirm the presence of macrolide resistance in M. genitalium in 18% of the infected patients. This figure is perfectly matching the findings from Dalarna County, a few hundred kilometres north of Stockholm, where the prevalence had increased to 18% in 2011,3 the year before the present study. However, it is significantly lower than prevalences around 40% reported from Denmark4 and the UK.7 We find it reasonable to believe that the difference in the prevalence of macrolide resistance between Denmark and Sweden with relatively similar healthcare systems could be due to differences in the use of azithromycin for treatment of chlamydia and urethritis of unknown aetiology. In Sweden, clinical guidelines recommend doxycycline as first line treatment for chlamydia and urethritis, whereas in Denmark, azithromycin 1 g single dose is almost exclusively used. Thus, we hypothesise that this observation may be due to the use of doxycycline as primary treatment of Chlamydia trachomatis.
An earlier infection with M. genitalium was significantly more common in patients carrying macrolide resistant M. genitalium strains. This may reflect that these patients, and probably also their partners, had been treated with azithromycin, and may have developed resistance during treatment. If a test of cure had not been meticulously carried out, asymptomatic infection may have persisted. Even more remarkable was the finding that treatment with azithromycin within the last 6 months was very strongly associated with macrolide resistance as 91% of patients having received azithromycin for any indication carried resistant strains compared with 5% of the azithromycin naïve. Unfortunately, a high proportion of the group of resistance negative patients had missing data. Macrolide resistance was found more often in women (24%) than in men (14%) although this difference did not reach statistical significance. The reason for this is not obvious, except that M. genitalium may more often be asymptomatic in women and use of macrolide treatment for other conditions may have selected for resistance. The emerging healthcare threat posed by azithromycin resistance in M. genitalium has challenged the efficacy of first line therapy and an increasing number of multidrug resistant M. genitalium strains have been reported from the Asia-Pacific region challenging second line treatment with moxifloxacin and leaving very few alternatives for treatment.1 At present, pristinamycin given as the maximal 4 g daily dose for 10 days appear to be the last, but not perfect solution.8 STI should be treated according to aetiology, and in the future, dual antimicrobial therapy may be needed to minimise the risk of development of antimicrobial resistance. It is also important to carry out a test of cure 3–4 weeks after start of azithromycin therapy since macrolide resistance develop during treatment and M. genitalium may be temporarily suppressed leading to a false negative test if performed earlier.9
Implementation of combined diagnostic and resistance tests for M. genitalium is urgently required alongside improved national and international surveillance to guide future treatment regimens.
Footnotes
Handling editor Jackie A Cassell
Contributors EB and CM collected clinical data and samples. JSJ supervised the laboratory testing. All authors contributed to the preparation of the manuscript.
Competing interests None declared.
Ethics approval Regional Ethics Committee in Stockholm, Karolinska Institutet, S-17177 Stockholm.
Provenance and peer review Not commissioned; externally peer reviewed.