Mycoplasma genitalium symptoms,concordance and treatment in high-risk sexual dyads

Abstract

The objective of this study was to determine the prevalence and concordance of Mycoplasma genitalium (MG) among Mexican American and African American women and their male sexual partners. Secondary objectives were to determine symptoms of MG infection and persistence of MG after antibiotic therapy. Heterosexual couples were tested for MG and interviewed separately regarding symptoms and behavioural/epidemiologic variables at baseline, six and 12 months. The overall prevalence of MG among women and men was 9.5% and 10.6%, respectively. Subjects were five times more likely to be infected with MG if their sexual partner was MG positive. Among men and women, MG prevalence and mean bacterial loads were similar after receiving single-dose azithromycin, doxycycline or no antibiotics. MG was associated with current urethral discharge in men. No clinical symptoms were specifically diagnostic of MG infection in women.

Keywords

concordance treatment efficacy genital symptoms

INTRODUCTION

Evidence continues to support that Mycoplasma genitalium (MG) is a sexually transmitted infection (STI) that may be asymptomatic^1–5 or cause non-gonococcal urethritis in men^2,6–11 and cervicitis,^3,4,12 endometritis¹³ and pelvic inflammatory disease^14–16 in women. These signs and symptoms have been shown to be independent of co-infection with Chlamydia trachomatis (CT) or Neisseria gonorrhoeae (GC).^12,16,17 Untreated MG results in persistent urethritis symptoms in men.^1,5,9,18 As an STI, MG is likely associated with fallopian tube damage. Women with tubal factor infertility are more likely to have antibodies to MG,¹⁹ and the presence of MG in men has been associated with low sperm concentration and abnormal sperm morphology.²⁰

Research has begun to focus on the most effective treatments for MG.^1,6,21–23 The regimens most commonly tested are therapies that are also effective for CT cervicitis: doxycycline (100 mg orally twice daily for seven days) or single-dose (1 g orally) azithromycin^1,11,18 and more recent trials examining single-dose rifalazil.²³ Evidence supports that these common antibiotic regimens are suboptimal for the treatment of MG, with eradication rates ranging from 17% to 82%.^{11,16,18,23,24} Extended azithromycin therapy (500 mg orally on day 1 followed by 250 mg orally daily on days 2–5), which is commonly used to treat Mycoplasma pneumoniae, has become the standard treatment for MG in Scandinavia,^1,21,22 where much of the initial screening and treatment trials have been conducted.

Few studies have been conducted in the USA on the prevalence of MG in the general population¹⁷ or among symptomatic patients.^23,24 It has been shown that MG is more prevalent in Afro-Caribbeans,¹¹ African Americans (AA) and Hispanics as compared with non-Hispanic whites,¹⁷ and thus data are needed on US minority populations on the prevalence, concordance and treatment of MG.

The objective of this study was to determine the prevalence and concordance of MG in high-risk sexual dyads enrolled in Project SAFE, a randomized controlled trial (RCT) of Mexican American (MA) and AA women with a non-viral STI and their male sexual partners, designed to prevent recurrent STIs. In addition, we determined symptoms associated with MG in women and men and tracked the efficacy of common treatments for non-viral STIs, specifically single-dose azithromycin and doxycycline, in eradicating MG.

MATERIALS AND METHODS

Participants and site

This study was approved by the Institutional Review Boards at the University of Texas Health Science Center at San Antonio and the San Antonio Metropolitan Health District. Briefly, MA and AA women, age 14–45 years old, diagnosed with a laboratory verified, non-viral STI including GC, CT, syphilis or Trichomonas vaginalis (TV) were referred to our research clinic. The women were informed prior to coming to our clinic that in order to participate in the study, they were required to invite their current male sexual partner (or a man with whom they have had sexual intercourse within the last 2 months) to the initial screening visit. Dyad enrollment occurred between 1 September 2005 and 1 June 2008. We screened 590 women and 572 men and subsequently enrolled 516 female–male dyads. At intake, men and women were interviewed separately, by a trained research assistant, specific to their gender. Dyads were then randomized to one of three groups: (1) individual control counselling for both; (2) behavioural intervention for the women and control counselling for the men; or (3) behavioural intervention for both (separate male and female groups). The control counselling lasted approximately 15 minutes and was provided by nurse clinicians or a physician assistant according to Centers for Disease Control guidelines.²⁵ The SAFE intervention entailed three, weekly, three-hour, small-group, multi-component behavioural cognitive interventions. We adapted the AIDS Risk Reduction Model to guide intervention development, supplemented with extensive ethnographic data to ensure suitability to our population.^26,27 At intervention sessions, we used role-playing, interactive video, handouts and group discussion to emphasize the preventive strategies of abstinence, periodic abstinence, mutual monogamy, correct and consistent use of condoms, full compliance with treatment protocols, reduction in the number of partners, avoidance of sexual intercourse until the woman and her partner(s) completed treatment, taking time between partners to be selective, avoidance of douching and seeking medical care whenever a woman suspected infection. Overall goals were to have women recognize their risk for contracting STIs, including HIV, commit to behaviour change and acquire the necessary skills to effect change. The primary outcome of Project SAFE was recurrent GC or CT. Because the objective of this study was not to report on the behavioural intervention, we included the samples on all patients who presented to our clinic, whether or not they subsequently enrolled in the behavioural intervention trial.

Laboratory detection of MG

All male and female participants were interviewed, examined, screened and treated for STIs at baseline, six and 12 months follow-up. Test of cures were performed 14–90 days after treatment (mean = 16 days). Men provided a casual (not first void) urine sample which was tested for GC, CT, TV and MG, and serum to test for syphilis. GC and CT testing were performed by polymerase chain reaction (PCR) and TV testing were performed using the TV InPouch system. We used the rapid plasma reagin test to screen for syphilis, and the fluorescent treponemal antibody test to confirm the diagnosis. Women had an endocervical swab obtained for GC, CT, TV and MG PCR testing, and serum to test for syphilis. Women also provided a non-first void urine sample for MG PCR testing. Detection of MG was performed by quantitative multiplex realtime PCR assays with inhibition control. Specific primers (MgPa-355F: 5′-GAGAAATACCTTGATGGTCAGCAA and MgPa-432R:5′-GTTAATATCATATAAAGCTCTACCGTTGTTATC) were designed to amplify the 76-bp region of the gene encoding cytadhesin MgPa, a major virulence factor, unique to MG. Generated amplicon was detected by the specific probe (MgPa-probe: 6FAM-ACTTTGCAATCAGAAGGT). Serial dilutions representing from one to 10⁸ genome copies of MG reference strain G37 (14th passage) were used to determine the efficiency and detection limit of the assay. We consistently detected five or more genomes per reaction and one genome per reaction with 60% probability. For quantification of MG in clinical specimens, five dilutions of G37 genomic DNA (ranging from one to 10⁷ genomes) were included as standard concentrations on each test plate. To evaluate the presence of inhibitory factors in the reaction, equal amounts of internal processing control were added to each reaction and detected by the second specific VIC-labelled probe. Multiplex TaqMan assays were performed in triplicates on 5 µL aliquots of DNA preparation in 50 µL reactions with Universal Master Mix as suggested by the manufacturer (Applied Biosystems, Foster City, CA, USA). Presented are average genome numbers per reaction. Individual aliquots represented 200 µL of urine or ∼1% of collected vaginal/endocervical swab. Assays were performed using ABI PRISM 7900HT Sequence Detection System. We defined very high MG burden as >100 genome copies/200 µL, which represented the 90th percentile of mean bacterial loads (MBLs) among male participants.

Definitions of male and female genital symptoms

We asked men if they had any of the following symptoms in the last 30 days: urethral discharge, sores, blisters, ulcers, itching, odour, pain and dysuria, increased urinary frequency, incomplete bladder emptying, fevers or chills. We used a composite scale to differentiate abnormal versus physiologic vaginal symptoms, which has been previously validated in our population and found to be associated with MG in a previous iteration of Project SAFE, which enrolled only women.^28,29 We asked women if they had vaginal discharge in the last 30 days and classified the discharge as ‘none’, ‘intermediate’ (meaning it was moderate or light in amount; red, brown, white or clear in colour; and curdy or watery in consistency) or ‘pathologic’ (meaning the vaginal discharge was heavy or ‘a lot’ in amount; yellow or green–gray in colour; slippery like mucus or creamy in consistency). We also asked women if the discharge caused them to take any of the following actions: stop you from having sex, go to a doctor, take medication for it, wear a pad or douche; we considered any ‘yes’ answer to indicate a vaginal discharge severe enough to warrant action.

Treatment of STIs

Subjects diagnosed with CT received azithromycin 1 g orally or doxycycline 100 mg twice daily for seven days. Patients diagnosed with TV received metronidazole 2 g orally. Participants diagnosed with GC initially (June 2005–August 2006) received ciprofloxacin 500 mg orally. Between August 2006 and April 2008 patients with GC received cefpodoxime 200 mg orally. Since May 2008, patients with GC have been given cefixime 400 mg orally. As there historically is no recommended antibiotic regimen for MG, patients did not receive medication specifically for MG infection.^30,31

Subjects were encouraged to return to our clinic for an unscheduled problem visit as needed for any symptoms of or concerns for re-infection. At each routine and problem visit a targeted physical examination was performed with collection of specimens for microbiological testing, including GC, CT, syphilis and TV. At all visits, participants were offered a test-of-cure for GC, CT, syphilis or TV following treatments and human immunodeficiency virus (HIV) testing. We provided STI treatment to the women and men enrolled in Project SAFE. We did not provide expedited partner treatment to additional sexual partners not enrolled in our study. However, we encouraged all participants diagnosed with an STI to inform all of their sexual partners of the diagnosis. Our research clinic is adjacent to the San Antonio Metropolitan Health District STI clinic, where sexual partners not enrolled in Project SAFE can obtain STI diagnosis and treatment.

Data analysis

Statistical analysis was performed using SPSS software (Chicago, IL, USA). Dichotomous variables were compared using χ ² statistic and continuous variables and means were compared using analysis of variance. Receiver operator characteristic (ROC) curves were used to correlate symptoms with MBLs of MG and MBLs were then stratified and correlated with genital symptoms.

RESULTS

Participants

At enrollment, all women had a laboratory verified, non-viral STI: 78.8% (465/590) had CT, 21.5% (127/590) had GC, 8.8% (52/590) had TV and 0.8% (5/590) had syphilis. We tested all male partners at enrollment: 11.0% (63/572) had GC, 39.9% (228/572) had CT, 0.5% (3/572) had TV and 0% (0/572) had syphilis. No participant entered the study with HIV but one male subject was diagnosed with HIV during the one-year trial. During the one-year study, at routine and problem visits, we collected 1037 urine specimens from men and have MG testing data on all 1037. We collected 1255 endocervical and 1255 urine specimens from women screened for enrollment in Project SAFE and have MG data for 1217 endocervical (39, 3.1% missing) and 1241 (15, 1.2% missing) urine samples.

Point prevalence and diagnostic test utility of MG

Table 1 details the prevalence and MBL of MG, based on visit type. Men had significantly higher MBLs than women (P < 0.001) and women had higher MBLs in endocervical compared with urine samples (P = 0.01). In women, the specificity of the urine MG test compared with the vaginal MG test was 93.9%, and the specificity of the endocervical MG test, as compared with the urine MG test, was 94.5%. The negative predictive value of the urine MG test in women was 94.5% and the negative predictive value of the endocervical MG test was 93.9%.

Table 1

Prevalence and mean bacterial load of Mycoplasma genitaliu m (MG) in men and women based on visit type

Gender and visit type	n	MG+ rate n (% visit type)	P value χ ²	Mean bacterial load (genome copies/ 200 µL) ± SD	P value ANOVA
Men ( n = 1037)
Intake	563	79 (14.0)	<0.001	754 ± 4131	0.007
6 months	262	20 (7.6)		226 ± 577
12 months	162	5 (3.1)		33 ± 55
Problem visits	50	6 (12.0)		10,154 ± 25,644
All routine visits (intake, 6 and 12 months)	987	104 (10.5)	0.74	68 ± 1213	<0.001
Problem visits	50	6 (12.0)	0.74	10,154 ± 25,644	<0.001
Women urine samples ( n = 1241)
Intake	584	75 (12.8)	<0.001	13 ± 38	0.58
6 months	329	25 (7.6)		9 ± 34
12 months	208	13 (6.3)		2 ± 5
Problem visits	120	5 (4.2)		26 ± 41
All routine visits (intake, 6 and 12 months)	1121	113 (10.1)	0.04	11 ± 34	0.36
Problem visits	120	5 (4.2)	0.04	26 ± 41	0.36
Women vaginal samples ( n = 1217)
Intake	581	62 (10.7)	0.07	238 ± 1086	0.64
6 months	324	30 (9.3)		52 ± 172
12 months	203	12 (5.9)		431 ± 1400
Problem visits	109	5 (4.6)		39 ± 64
All routine visits (intake, 6 and 12 months)	1108	104 (9.4)	0.09	206 ± 964	0.70
Problem visits	109	5 (4.6)	0.09	39 ± 64	0.70

ANOVA = analysis of variance

Concordance of MG among heterosexual dyads

Table 2 details the concordance of MG among sexual dyads. By definition these data include only results of male and female dyadic samples obtained at the same visit. Among all visits, men and women were approximately five times more likely to be infected with MG if their sexual partner was currently infected with MG. We compared dyad data obtained at the intake (when all women were currently or recently infected with a non-viral STI) visit versus the six- and 12-month follow-up visits to determine whether the likelihood of MG concordance changed when the women were co-infected with another STI. The odds of MG concordance were higher at the intake visit (OR = 5.27) than at the six- and 12-month follow-up visits (OR = 2.88).

Table 2

Concordance* of Mycoplasma genitalium (MG) among high-risk sexual dyads

Index men or women and visit type	Concordance: sexual partner is MG positive at the same visit n (n/row total%)	OR (95% CI)
Men all routine visits	Women is MG +
MG positive (n = 89)	36 (40.4)	4.98 (3.10, 7.98)
MG negative (n = 791)	95 (12.0)	4.98 (3.10, 7.98)
Men initial visit only	Women is MG +
MG positive (n = 69)	31 (44.9)	5.27 (3.05, 9.11)
MG negative (n = 425)	57 (13.4)	5.27 (3.05, 9.11)
Men 6- and 12-month follow-up visits only	Women is MG +
MG positive (n = 20)	5 (25.0)	2.88 (1.03, 8.08)
MG negative (n = 366)	38 (10.4)	2.88 (1.03, 8.08)
Women all routine visits	Men is MG +
Urine and/or vaginal MG positive (n = 131)	36 (27.5)	4.98 (3.10, 7.98)
Urine and vaginal MG negative (n = 749)	53 (7.1)	4.98 (3.10, 7.98)
Women initial visit only	Men is MG +
Urine and/or vaginal MG positive (n = 88)	31 (35.2)	5.27 (3.05, 9.11)
Urine and vaginal MG negative (n = 406)	38 (9.4)	5.27 (3.05, 9.11)
Women 6- and 12-monthfollow-up visits only	Men is MG +
Urine and/or vaginal MG positive (n = 43)	5 (11.6)	2.88 (1.03, 8.08)
Urine and vaginal MG negative (n = 343)	15 (4.4)	2.88 (1.03, 8.08)

*Data described are for a female and male dyad sample obtained at the same visit

Persistence of MG after antimicrobial therapy

Table 3 details the prevalence and MBLs of MG, with attention to levels of antimicrobial therapy received in the last 30 days. The antibiotic therapy was prescribed for the treatment of CT, not specifically for MG. Among men and women, none of the standard therapies for uncomplicated CT infection (single-dose azithromycin or doxycycline) or antibiotics recorded as ‘other’ for uncomplicated TV or GC were significantly better at eliminating MG than no antimicrobial therapy.

Table 3

Prevalence and mean bacterial load of Mycoplasma genitaliu m (MG) in men's and women's visits based on antimicrobial use within the past 30 days

Gender and antimicrobials taken in past 30 days	n	MG+ rate n (row %)	P value χ ²	MG bacterial load (genome copies/200 µL) mean ± SD	P value ANOVA
Men ( n = 1037)
None	967	105 (10.9)	0.58	1174 ± 6858	0.96
Azithromycin	49	4 (7.6)		464 ± 667
Doxycycline	8	1 (3.1)		0.8
Other	13	0 (0.0)		–
Women urine samples ( n = 1206)
None	1037	91 (8.8)	0.10	13 ± 38	0.50
Azithromycin	111	12 (10.8)		2 ± 3
Doxycycline	14	0 (0)		5 ± 7
Other	44	8 (18.2)		11 ± 35
Women vaginal samples ( n = 1268)
None	1037	91 (8.8)	0.97	141 ± 655	<0.001
Azithromycin	110	9 (8.2)		8 ± 15
Doxycycline	13	1 (7.7)		1
Other	108	3 (7.1)		2572 ± 4455

Genital symptoms of MG in men and women

Table 4 details the presence or absence of genital symptoms among men and women with MG and provides subgroup analysis of whether the subjects were co-infected with GC or CT. We focused on urethral discharge as the male symptom because analysis of the following symptoms among MG+ men with high bacterial loads: sores/blisters/ulcers (n = 2), dysuria (n = 6), increased urinary frequency (n = 9), incomplete bladder emptying (n = 6), fevers/chills (n = 5), penile itching (n = 5), penile odour (n = 4), painful penis (n = 4) were too infrequent to make meaningful statistical analysis possible. Among men, MG infection was significantly associated with current urethral discharge. The association persisted among men with MG infection alone, but was confounded by GC and CT co-infection. Using ROC curves, we found that even low MBLs of MG were able to differentiate urethritis symptoms in men. Among men without CT or GC co-infection at the time of sampling (n = 787), 23.8% (5/21) with MBL MG counts of 20 genomes/200 µL or greater reported current urethral discharge; 2.0% (1/49) with MBL MG counts of >0.0–19.99 genomes/200 µL reported current urethral discharge and 3.1% (22/717) with 0 MBL MG reported current urethral discharge. For women, we did not find a significant association between the complaint of intermediate or pathological vaginal discharge or actions taken in response to the presence of vaginal discharge and MG infection with or without GC and or CT co-infection.

Table 4

Association of Mycoplasma genitaliu m (MG) and genital symptoms among subjects with and without CT and or GC co-infection*

	MG status				χ ² P value	OR (95% CI)
	Positive		Negative
Symptom reported at visit	n	Symptom + n (%)	n	Symptom + n (%)
Men penile discharge present
Yes, now
All men	102	11 (10.8)	889	39 (4.4)	0.01	2.64 (1.32, 5.27)
No CT or GC	68	6 (8.8)	716	21 (2.9)	0.02	3.23 (1.28, 8.02)
CT and/or GC	34	5 (14.7)	173	18 (10.4)	0.47	1.49 (0.53, 4.19)
Yes, gone now
All men	102	9 (8.8)	889	67 (7.5)	0.69	1.19 (0.58, 2.43)
No CT or GC	68	6 (8.8)	716	58 (8.1)	0.82	1.10 (0.47, 2.59)
CT and/or GC	34	3 (8.8)	173	9 (5.2)	0.42	1.76 (0.49, 6.42)
Women vaginal discharge present (MG from vaginal sample)
Intermediate vaginal discharge
All women	102	23 (22.5)	971	255 (26.3)	0.48	0.82 (0.51, 1.32)
No CT or GC	70	15 (21.4)	742	188 (25.3)	0.47	0.80 (0.45, 1.45)
CT and/or GC	32	8 (25.0)	229	67 (29.3)	0.62	0.81 (0.35, 1.85)
Pathological vaginal discharge
All women	102	20 (19.6)	971	236 (24.3)	0.29	0.76 (0.46, 1.26)
No CT or GC	70	10 (14.3)	742	164 (22.1)	0.13	0.59 (0.30, 1.16)
CT and/or GC	32	10 (31.3)	229	72 (31.4)	0.98	0.99 (0.45, 2.17)
Took action due to vaginal discharge
All women	102	36 (35.3)	971	411 (42.3)	0.17	0.74 (0.49, 1.14)
No CT or GC	70	22 (31.4)	742	290 (39.1)	0.21	0.71 (0.42, 1.20)
CT and/or GC	32	14 (43.8)	229	121 (52.8)	0.34	0.69 (0.33, 1.45)

GC = Neisseria gonorrhoeae; CT = Chlamydia trachomatis

*Data described include when symptoms of MG, GC and CT data were available for each sampling

There were 23 men with very high (>100 genome copies/200 µL) bacterial loads (range 324–60,459). None were homosexual or bisexual. One man reported 200 lifetime sexual partners and one reported 110 lifetime sexual partners. The mean number of lifetime sexual partners among the 23 men with high MG loads was 26 and the median was 15. Two of these 23 men had received azithromycin 1 g orally within the last 30 days. Among the 23 men with very high counts, 39% (9/23) were co-infected with CT, 13% (3/23) were infected with GC and none were infected with TV or syphilis. The MBL among these 23 men with high counts was not statistically different among those with and without CT co-infection (P = 0.13). Similarly, the MBL among these 23 men was similar with and without GC co-infection (P = 0.51). Among the 23 men with high bacterial loads: 22% (5/23) had urethral discharge. In this cohort, we have concordancy data on 11 dyads at 24 separate measurement intervals. At 10 intervals, the female partner was MG negative, despite high MBLs in the man. At five intervals, the men and the women were both positive for MG and at 14 intervals, both partners were negative for MG. Of note, the man with the highest bacterial load, 60,457 genome copies/200 µL, was assessed at a problem visit for which we do not have data on his female partner. The man with the second highest bacterial load, 35,202 genome copies/200 µL, had an infected female partner, with a bacterial load of 18 genome copies/200 µL.

Among women, the average MBL for urine samples was 0.98 ± 10 genome copies/200 µL and was 17.80 ± 287 genome copies/200 µL for endocervical samples. The maximum count for endocervical samples was 7718 genome copies/200 µL and 292 genome copies/200 µL for female urine samples.

DISCUSSION

We found that among high-risk sexual dyads, the overall prevalence of MG in men was 10.6%, which was similar to the prevalence of GC in men at intake (11.0%) and is in the range of previous studies with high-risk populations.^6,8 The prevalence of MG among women enrolled in Project SAFE was 9.5% (118/1241 urine samples) and 9.0% (109/1216 vaginal samples). We demonstrated concordance of MG among sexual dyads and the likelihood of concordance was higher at the intake visit, when all women, by definition, were currently co-infected with a non-viral STI. None of the standard therapies for uncomplicated CT infection were significantly better than no therapy at eliminating MG. Finally, MG was associated with urethral discharge in men not co-infected with another STI, but no symptom was diagnostic of MG in women.

Among men enrolled in Project SAFE, the highest prevalence and MBLs of MG were detected at problem visits. We found that MG, without CT or GC co-infection, was significantly associated with urethritis symptoms in men, in agreement with previous studies that have established urethritis as a symptom of MG in men.^6–11,32 Using ROC curves, we found that MG MBLs of 20 genome copies/200 µL or more could differentiate current urethritis symptoms in men not co-infected with GC or CT. In addition, we describe 23 men with very high MBLs of MG (>100 genome copies/200 µL urine). The most common complaint in this subset was urethral discharge.

There is debate as to whether MG causes symptoms in women.^2–5,29 Female MG infection is described as either asymptomatic or similar to CT symptoms by some.^1–5,17,29 Using a validated composite symptom scale,²⁸ we found that vaginal discharge or taking action due to the presence of vaginal discharge was not predictive of MG infection, which parallels the difficulty in determining whether symptoms in women are physiologic or pathologic.^28,31 In a previous Project SAFE cohort, which recruited only women, we found that MG was associated with intermediate vaginal discharge and action in response to the discharge.²⁹ However, the MG testing performed in this prior study was a mixture of culture, PCR and enzyme linked immunosorbent assay.²⁹ In addition, there was a high rate of bacterial vaginosis in this previous SAFE cohort, and once the analysis of symptoms was adjusted for other co-infections, the association of MG and vaginal discharge was not as strong.²⁹ It is unclear if different MG strains have markedly different virulence properties in terms of host cell interaction and severity of tissue damage.³³ Among women, the highest MG prevalence was at intake and the lowest prevalence of MG and the lowest MBLs were found at problem visits.

It is recommended that MG PCR be performed with urine from men and urine and endocervical/vaginal swabs from women.⁵ We found significantly higher MG MBLs in men than in women. This may be due to several factors including urethral length, degree of virulence, tissue tropism, STI co-infection, immunologic response and genetic factors.³⁴ We also noted that in women, MG MBLs in endocervical specimens were higher than in urine specimens. This is consistent with MG being an endocervical/vaginal pathogen in women rather than a urinary pathogen. In addition, we have previously shown that MG is able to reside intracellularly in human genital cells.³⁴ However, like urine GC and CT PCR probes,³¹ the specificity of urine or vaginal samples was similar.

We found that men and women were approximately five times more likely to be infected with MG if their sexual partner was currently infected with MG. Several other studies have described the concordance rate of MG among heterosexual couples (range 5–66%).^{2,4,7,8,35,36} However, all of these studies are limited by smaller cohorts (n = 2–39) of sexual dyads.^{2,4,7,8,35,36} In addition, we found that the likelihood of MG concordance was higher at the intake visit (OR = 5.27) when all women, by definition, had a current or recent STI co-infection versus the odds of concordance at the six- and 12-month follow-up visits (OR = 2.88). There are several studies which show that HIV infectivity is enhanced by the presence of other STIs.^31,37 Manhart et al. ³⁸ found that MG positivity was associated with higher shedding rates of HIV-1 DNA from the cervix.

It has been established in several trials that doxycycline and single-dose azithromycin therapy have suboptimal eradication rates for MG (17–82%)^{1,6,7,11,24,39,40} and the majority of men with treatment failure have persistent urethritis symptoms.^6,7,10,40 Björnelius et al. ¹ in an RCT, established that extended azithromycin therapy, which is commonly used for the treatment of M. pneumoniae was the most effective treatment for MG. This extended azithromycin therapy, which has been shown to have 96–99% efficacy at 20–56 day follow-up, has become the standard treatment for MG in Scandinavia and Australia.^1,7,22,39,40 We found that men and women receiving antibiotics for common STIs within the last 30 days had similar prevalence and MBLs of MG as subjects receiving no antibiotic therapy. Of note, the highest MBLs among men were found in those receiving no antibiotic therapy and were found among women receiving medications other than azithromycin or doxycycline. This may suggest that azithromycin and doxycycline (as compared with none and other meds) had some eliminating effect on MG, resulting in a trend, although not significant, toward lower bacterial burdens. We decided to examine antibiotic exposure within the last 30 days because many of the treatment trials perform test-of-cure at 30 days, or ask about symptoms within the last 30 days.^1,8,10,23

Strengths of our study include that we describe prevalence, concordance, MBL and treatment data on a large cohort of high-risk sexual dyads and all women initially entered the trial with a laboratory verified STI. We also describe a small cohort of men with very high MG bacterial loads, which has not been previously reported. Our data are limited by the fact that this was not an MG treatment trial.

MG was as common among high-risk men as GC. MG was associated with penile discharge in men, but was difficult to diagnose, based on clinical symptoms in women. MG persisted after receiving single-dose azithromycin or doxycycline. This large data-set provides further evidence that MG is an STI, which is not adequately treated with other common STI therapies. We plan to prospectively study the persistence of MG in dyads after extended azithromycin therapy. In addition, we will use realtime PCR and confocal immunoanalysis to determine the location of MG in endocervical cells if treatment failure occurs and cannot be explained by repeat exposure.

Footnotes

ACKNOWLEDGEMENT

This study was supported by a grant (U01 AI200029) from the National Institute of Allergy and Infectious Diseases.

References

Björnelius

, Anagrius

, Bojs

, Antibiotic treatment of symptomatic Mycoplasma genitalium infection in Scandinavia: a controlled clinical trial. Sex Transm Inf 2008;84:72–6

Tosh

, Van der Pol

, Fortenberry

, Mycoplasma genitalium among adolescent women and their partners. J Adolesc Health 2007;200:412–17

Manhart

, Critchow

, Holmes

, Mucopurulent cervicitis and Mycoplasma genitalium . J Infect Dis 2003;187:650–7

Falk

, Fredlund

, Jensen

. Signs and symptoms of urethritis and cervicitis among women with or without Mycoplasma genitalium or Chlamydia trachomatis infection. Sex Transm Infect 2005;81:73–8

Stein

, Baseman

. The evolving saga of Mycoplasma genitalium . Clin Microbiol Newslett 2006;28:41–8

Jensen

. Mycoplasma genitalium: the aetiological agent of urethritis and other sexually transmitted diseases. J Eur Acad Dermatol Venereol 2004;18:1–11

Wikström

, Jensen

. Mycoplasma genitalium: a common cause of persistent urethritis among men treated with doxycycline. Sex Transm Inf 2006;82:276–9

Falk

, Fredlund

, Jensen

. Symptomatic urethritis is more prevalent in men infected with Mycoplasma genitalium than with Chlamydia trachomatis . Sex Transm Infect 2004;80:289–93

Yokoi

, Maeda

, Kubota

, The role of Mycoplasma genitalium and Ureaplasma urealyticum Biovar 2 in postgonoccocal urethritis. Clin Infect Dis 2007;45:866–71

10.

Horner

, Thomas

, Gilroy

, Egger

, Taylor-Robinson

. Role of Mycolplasma genitalium and Ureaplasma urealyticum in acute and chronic nongonococcal urethritis. Clin Infect Dis 2001;32:995–1003

11.

Horner

, Gilroy

, Thomas

, Naaidoo

ROM

, Taylor-Robinson

. Association of Mycoplasma genitalium with acute non-gonococcal urethritis. Lancet 1993;342:582–5

12.

Pepin

, Labbe

, Khonde

, Mycoplasma genitalium: an organism commonly associated with cervicitis among west African sex workers. Sex Transm Infect 2005;81:67–72

13.

Cohen

, Manhart

, Bukusi

. Association between Mycoplasma genitalium and acute endometritis. Lancet 2002;359:765–6

14.

Simms

, Eastick

, Mallinson

, Associations between Mycoplasma genitalium, Chlamydia trachomatis, and pelvic inflammatory disease. Sex Transm Infect 2003;79:154–6

15.

Cohen

, Mugo

, Astete

, Detection of Mycoplasma genitalium in women with laparoscopically diagnosed acute salpingitis. Sex Transm Infect 2005;81:463–6

16.

Haggerty

. Evidence for a role of Mycoplasma genitalium in pelvic inflammatory disease. Curr Opin Infect Dis 2008;21:65–9

17.

Manhart

, Holmes

, Hughes

, Houston

, Totten

. Mycoplasma genitalium among young adults in the United States: an emerging sexually transmitted infection. Am J Public Health 2007;97:1118–25

18.

Johannisson

, Enstrom

, Lowhagen

, Occurrence and treatment of Mycoplasma genitalium in patients with urethritis. Int J STD AIDS 2000;11:324–6

19.

Clausen

, Fedder

, Drasbek

, Serological investigation of Mycoplasma genitalium in infertile women. Hum Reprod 2001;16:1866–74

20.

Gdoura

, Kchaou

, Chaari

, Ureaplasma urealyticum, Ureaplasma parvum, Mycoplasma hominis and Mycoplasma genitalium infections and semen quality of infertile men. BMC Infect Dis 2007;7:129–37

21.

Falk

, Fredlund

, Jensen

. Tetracycline treatment does not eradicate Mycoplasma genitalium . Sex Transm Infect 2003;79:318–19

22.

Ross

JDC

, Jensen

. Mycoplasma genitalium as a sexually transmitted infection: implications for screening, testing, and treatment. Sex Transm Infect 2006;82:269–71

23.

Stamm

, Batteiger

, McCormack

, A randomized, double-blind study comparing single-dose rifalazil with single-dose azithromycin for the empirical treatment of nongonococcal urethritis in men. Sex Transm Dis 2007;34:545–52

24.

Haggerty

, Totten

, Astete

, Ness

. Mycoplasma among women with nongonococcal, nonchlamydial pelvic inflammatory disease. Infect Dis Obstet Gynecol 2006;2006:30184

25.

Centers for Disease Control and Prevention. HIV prevention through early detection and treatment of other sexually transmitted diseases – United States. MMWR 1998;47 (RR-12):1–24

26.

Catania

, Kegeles

, Coates

. Towards an understanding of risk behavior: an AIDS risk reduction model (ARRM). Health Educ Q 1990;17:53–72

27.

Ramos

, Shain

, Johnson

. ‘Men I mess with don't have anything to do with AIDS’: using ethno-theory to understand sexual risk perception. Sociol Q 1995;36:483–504

28.

Piper

, Korte

, Holden

AEC

, Development of composite symptom variables for quantitative analysis of genitourinary symptomatology in women. Int J STD AIDS 2005;16:128–32

29.

Korte

, Baseman

, Cagle

, Cervicitis and genitourinary symptoms in women culture positive for Mycoplasma genitalium . Am J Reprod Immunol 2006;55:265–75

30.

Frenkl

, Potts

. Sexually transmitted infections. Urol Clin N Am 2008;35:33–46

31.

Centers for Disease Control and Prevention. 2006 sexually transmitted diseases treatment guidelines. MMWR 2006;55 (RR 1–11):1–94

32.

Leung

, Eastick

, Haddon

, Horn

, Ahuja

, Horner

. Mycoplasma genitalium is associated with symptomatic urethritis. Int J STD AIDS 2006;17:285–8

33.

Ueno

, Timenetsky

, Centonze

, Interaction of Mycoplasma genitalium with host cells: evidence of nuclear localization. Microbiol 2008; 154:3033–41

34.

Blaylock

, Musatovova

, Baseman

. Determination of infectious load of Mycoplasma genitalium in clinical samples of human cells. J Clin Microbiol 2004;42:746–52

35.

Keane

FEA

, Thomas

, Gilroy

, The association of Chlamydia trachomatis and Mycoplasma genitalium with non-gonococcal urethritis: observations on heterosexual men and their female partners. Int J STD AIDS 2000;11:435–9

36.

Hjorth

, Björnelius

, Lidbrink

, Sequence-based typing of Mycoplasma genitalium reveals sexual transmission. J Clin Microbiol 2006;44:2078–83

37.

Baeten

, Overbaugh

. Measuring the infectiousness of perons with HIV-1: opportunities for preventing sexual HIV-1 transmission. Curr HIV Res 2003;1:69–86

38.

Manhart

, Mostad

, Baeten

, Astete

, Mandaliya

, Totten

. High Mycoplasma genitalium organism burden is associated with shedding of HIV-1 DNA from the cervix. J Infect Dis 2008;197:733–6

39.

Carlberg

, Bjornelius

, Jensen

. Mycoplasma genitalium – the search for effective treatment. Int J STD AIDS 2002;13(Suppl. 1):30–1

40.

Bradshaw

, Jensen

, Tabrizi

, Azithromycin failure in Mycoplasma genitalium urethritis. Emerg Infect Dis 2006;12:1149–52