Objective
We sought to analyze the prevalence and clinical manifestations of Mycoplasma genitalium infection in a heterogeneous population of women.
Study Design
The study was designed as a cross-sectional case-control study. Women attending a gynecological outpatient service from 2003 through 2008 were invited to participate.
Results
The prevalence of M genitalium was 2.1% and of Chlamydia trachomatis was 2.8% among 5519 tested women. A total of 679 women were included. Both pelvic inflammatory disease (PID) and cervicitis were independently associated with M genitalium (odds ratio, 9.00; 95% confidence interval, 1.62–49.89 and odds ratio, 3.80; 95% confidence interval, 2.06–7.03, respectively). Women with C trachomatis had a higher frequency of both PID (18.3% vs 4.9%, P < .001) and cervicitis (33.4% vs 22.3%, P < .001) than women with M genitalium .
Conclusion
M genitalium was an independent and strong risk factor for both cervicitis and PID although, compared to C trachomatis , clinical manifestations were less frequent.
Numerous studies have associated Mycoplasma genitalium with nongonococcal urethritis in men and demonstrated its importance in this disease. M genitalium is also recognized as a sexually transmitted agent as demonstrated in several studies by a high concordance rate of infection between partners and by molecular typing.
See Journal Club, page 532
While M genitalium is well documented as an agent of nongonococcal urethritis in men, the manifestations of M genitalium infection in women are less well described. Although an association between M genitalium and symptoms of lower genital tract inflammation including cervicitis has been reported in several studies there are other studies in which no such association was found.
The role of M genitalium in more severe infections such as pelvic inflammatory disease (PID) has been investigated but data are still limited. The association with PID has been shown by endometrial biopsy specimens in women with acute endometritis, in a single Fallopian tubal specimen in a HIV-positive woman with acute salpingitis, and in one case–control study in women with clinically diagnosed PID. Our group has previously published a case–control study showing M genitalium is an independent risk factor for PID after termination of pregnancy.
Recently a prospective cohort study reported on the prevalence of M genitalium and its role in PID among female students in London, United Kingdom. At 12 months of follow-up, the incidence of PID in M genitalium –positive women was 3.9% compared to 1.7% among M genitalium –negative women, but the difference was not statistically significant.
Although there is some evidence for the role of M genitalium in cervicitis and PID, results from earlier studies have mainly comprised patients attending units for sexually transmitted infections (STI) or populations with a high level of HIV-infected women. In this study we investigated a large urban population of women presenting with various gynecological symptoms not necessarily connected with STI. The purpose of the present study was to analyze the prevalence, symptoms, and clinical manifestations of M genitalium infection in a heterogeneous population of women, including women in early pregnancy.
Materials and Methods
Study population
From November 2003 through June 2008 women attending the emergency gynecological outpatient service at Skane University Hospital in Malmö, Sweden, were invited to participate. The study was designed as a case–control study. Women with various gynecological symptoms of acute or semiacute nature, including women with pathological symptoms in early pregnancy, constituted the study population. All women were examined by an experienced specialist in gynecology.
Women visiting the outpatient clinic are routinely offered screening for Chlamydia trachomatis . In women participating in the study, the C trachomatis sample was later used for M genitalium detection. The results of M genitalium testing were not disclosed until the study was completed.
Patients with C trachomatis infection were treated with doxycycline (100 mg), twice on the first day, and then with 100 mg for another 8 days. Pregnant women received treatment with amoxicillin, 500 mg 3 times a day for 7 days.
Cases consisted of M genitalium –positive and C trachomatis –positive women. Controls testing negative for C trachomatis and M genitalium infection during the same month were chosen randomly. They were age matched within a range of 12 months. The local ethics committee gave approval to the study. Informed consent was obtained from all participants.
Data collection
A protocol with variables such as gynecological characteristics, self-reported symptoms, clinical findings, and clinical diagnoses was designed and data on cases and controls were collected from medical records after the study was closed.
Criteria for the clinical diagnosis of cervicitis and PID were in agreement with the guidelines from the Centers for Disease Control and Prevention, Atlanta, GA, and the diagnosis of bacterial vaginosis was based on the criteria of Amsel et al.
The clinical diagnosis of cervicitis was based on either pathological saline-prepared vaginal wet smears (more leukocytes than epithelial cells in the absence of clue cells and/or inflammatory vaginitis) or pathological cervical discharge (pus from the endocervical canal) or friability together with cervical motion tenderness.
The clinical diagnosis of PID was based on lower abdominal pain together with cervical and/or uterine and/or adnexal tenderness at pelvic bimanual examination and one of the following signs: pathological saline-prepared vaginal wet smear (more leukocytes than epithelial cells in the absence of clue cells and inflammatory vaginitis) or pus from the endocervical canal or elevated C-reactive protein or fever (oral temperature >38.0°C).
Microbiological analyses
From 2003 first void urine together with cervical samples were obtained for C trachomatis testing and from 2005, urine together with a vaginal swab. During the study period 2 different tests were used for C trachomatis . From the start of the study until mid-2006 the Cobas Amplicor (Roche Molecular Diagnostics, Pleasanton, CA) was used. It was replaced by the m2000 (Abbott Molecular Inc., Des Plaines, IL) during the later part of the study period.
In 2006, a new variant of C trachomatis was discovered in Sweden with a deletion in the cryptic plasmid, which contained the target sequences for both these tests. The new variant of C trachomatis constituted 25-30% of the cases in 2006 and must have appeared after 2000/2001 when no such strains could be detected among 258 culture-positive cases tested retrospectively. From 2007 the m2000 test has been modified to detect the new variant.
Samples obtained for C trachomatis testing were also used to test for M genitalium . Stored samples were pooled with 5-8 samples in each pool and then submitted to polymerase chain reaction (PCR). Positive pools were then resolved by testing the individual samples of the positive pool separately.
M genitalium was detected using the PCR and targeting the operon of the surface adhesion protein MgPa. The method used has been well described. To increase sensitivity and specificity, a seminested PCR was used where the primer MgPa-1 was used together with MgPa-3 in the first step while the MgPa-1 primer was combined with MgPa-2 in the second step. The amplification product was identified after gel electrophoresis and staining by ethidium bromide. The identity of the strains detected in this way was later confirmed by DNA sequencing. The method has been compared by others to a TaqMan PCR developed on the same genetic target with a sensitivity of <5 genomic copies.
Testing for Neisseria gonorrhoeae was not included in the opportunistic screening at the gynecological outpatient service in Malmö as N gonorrhoeae is very rare in Sweden with an incidence of 6.6-7.8/100,000 from 2003 through 2008. Testing was performed in selected cases when genital infection was suspected.
Statistical analysis
Data were analyzed using SPSS for Windows version 16.0 (SPSS Inc, Chicago, IL). Frequencies of gynecological characteristics, symptoms, clinical signs, and diagnoses (categorical variables) were compared among M genitalium –positive patients, C trachomatis –positive patients, M genitalium –negative controls, and C trachomatis –negative controls using the Pearson χ 2 test and Fisher exact test when small numbers occurred. Backward stepwise binary logistic regression was used to adjust for possible confounders such as age. A 95% confidence interval (CI) was calculated.
Results
Prevalence of M genitalium and C trachomatis
A majority of the invited women, 89.6% (5519/6159) accepted participation in the study. In all, 5519 patients were tested for both M genitalium and C trachomatis during the study period. M genitalium infection was demonstrated in 2.1% (116/5519) and C trachomatis infection in 2.8% (149/5519) of the tested patients. Dual infection was noted in 3.8% (10/265).
Medical records from 106 M genitalium –positive and 138 C trachomatis –positive patients were retrieved and included in the study together with 451 negative controls. Among these, 8 patients were infected by both M genitalium and C trachomatis . They were excluded from Pearson χ 2 test and Fisher exact test analyses. There where 1.97 controls/case. During the whole study period 1390 patients were tested for N gonorrhoeae . No single case was detected.
Age and gynecological characteristics
The mean age was 25.7 (15-46) years in patients with M genitalium infection and was 26.5 (16-52) years in patients with C trachomatis infection. The age distribution showed that younger women had higher rates of both C trachomatis and M genitalium . The peak for C trachomatis infection was seen in women <20 years of age with declining rates after that. The highest rates of M genitalium were observed slightly later in women of 20-24 years of age with similar rates as C trachomatis up to 35 years of age ( Figure 1 ) .
The pregnancy rate of the controls was 26.6%. There was no difference in pregnancy rate between M genitalium –positive women (13.3%) and C trachomatis –positive women (9.9%) and both M genitalium –positive and C trachomatis –positive women were less likely to be pregnant ( P = .005 and P = .001) than the controls.
There was no difference in the use of contraception between M genitalium –positive women and controls. Current use of any contraception was reported more often in C trachomatis –positive women than negative controls but there was no difference in the use of hormonal contraception or barrier methods (condoms). The use of intrauterine copper device was higher in C trachomatis –positive women compared to both M genitalium –positive women and controls ( Table 1 ).
| Gynecological characteristics | M genitalium –positive cases, all = 98 | C trachomatis –positive cases, all = 130 | Negative controls, all = 451 | P value, M genitalium / C trachomatis positive vs negative controls | P value, M genitalium vs C trachomatis positive |
|---|---|---|---|---|---|
| Mean age, y (range) | 25.7 (15–46) | 26.1 (16–60) | 26.5 (15–60) | .77/.28 | .42 |
| Current pregnancy | 13/98 (13.3) | 12/121 (9.9) | 118/443 (26.6) | .005/< .001 | .57 |
| Previous pregnancy | 52/86 (60.5) | 55/100 (52.3%) | 223/376 (59.3) | .26/.26 | .55 |
| Proliferative phase in menstrual cycle (d 6-14) | 14/53 (26.4) | 16/56 (28.6) | 49/159 (30.8) | .54/.75 | .97 |
| Current use of any contraception | 36/89 (40.4) | 58/109 (49.1) | 164/398 (41.2) | .90/.02 | .53 |
| Hormonal contraception a | 30/89 (33.7) | 33/111 (29.7) | 70/232 (30.1) | .54/.93 | .65 |
| IUD (copper) | 5/89 (5.6) | 19/110 (17.3) | 39/397 (9.8) | .31/.03 | .02 |
| Barrier | 2/89 (2.2) | 2/110 (4.5) | 14/397 (3.5) | .54 b /.62 b | .61 b |
a Combined hormonal contraception and gestagen methods;
Clinical diagnoses
After exclusion of dual infections, PID was diagnosed in 4.9% (4/81) of women with M genitalium infection and in 0.6% (2/317) of the controls ( P = .01) ( Table 2 ). An independent association between M genitalium and PID was also seen in logistic regression after adjustment for age and C trachomatis (odds ratio [OR], 9.00; 95% CI, 1.62–49.89) ( Table 3 ). Condom use was reported in 4 cases and in 14 controls. The differences in PID remained statistically significant after excluding these patients as well from the logistic regression analysis.
| Characteristics | M genitalium –positive cases, all = 98 (%) | C trachomatis –positive cases, all = 130 (%) | Negative controls, all = 451 (%) | P value, M genitalium positive vs negative controls | P value, C trachomatis positive vs negative controls | P value, M genitalium vs C trachomatis positive |
|---|---|---|---|---|---|---|
| Clinical diagnose | ||||||
| Pelvic inflammatory disease | 4/81 (4.9) | 20/109 (18.3) | 2/346 (0.6) | .013 b | < .001 b | .006 |
| 4/81 (4.9) a | 20/106 (18.9) a | 2/317 (0.6) a | .013 a , b | .001 a , b | .005 a | |
| Cervicitis | 21/94 (22.3) | 45/117 (38.4) | 30/429 (7.0) | < .001 | < .001 | .01 |
| 21/81 (25.9) a | 45/105 (42.8) a | 29/316 (9.2) a | < .001 a | < .001 a | .02 a | |
| Bacterial vaginosis | 14/94 (14.9) | 15/117 (12.8) | 38/429 (8.8) | .08 | .20 | .66 |
| 14/81 (17.3) a | 14/105 (13.3) a | 33/315 (10.5) a | .13 a | .54 a | .46 a | |
| Vulvovaginal candidiasis | 4/94 (4.2) | 3/118 (2.5) | 28/414 (6.8) | .36 | .08 b | .49 b |
| 4/81 (4.9) a | 3/105 (2.8) a | 28/312 (9.0) a | .24 a | .07 a | .46 a | |
| Clinical sign | ||||||
| Pathological vaginal wet smear (microscopy) | 24/43 (55.8) | 35/60 (58.3) | 83/178 (46.6) | .28 | .12 | .80 |
| 24/42 (57.1) a | 34/58 (58.6) a | 79/162 (48.8) a | .33 a | .14 a | .88 a | |
| 11/27 (40.7) c | 23/44 (52.3) c | 39/126 (31.0) c | .33 c | .03 c | .28 c | |
| Friability | 4/90 (4.4) | 12/114 (10.5) | 11/411 (2.7) | .30 | < .001 | .11 |
| 4/78 (5.1) a | 12/101 (11.8) a | 8/302 (2.6) a | .26 a | < .001 a | .12 a | |
| Cervical tenderness | 20/93 (21.5) | 45/111 (40.5) | 60/407 (14.7) | .10 | < .001 | .001 |
| 19/80 (23.8) a | 43/98 (43.9) a | 54/298 (18.1) a | .26 a | < .001 a | .005 a | |
| C-reactive protein >8, mg/L | 9/53 (17.0) | 25/82 (30.5) | 43/279 (15.4) | .77 | .02 | .08 |
| 9/52 (17.3) a | 24/76 (31.6) a | 40/227 (17.6) a | .96 a | .01 a | .07 a | |
| Fever (>38.0°C) | 2/73 (2.7) | 15/71 (21.1) | 13/283 (4.6) | .38 b | .001 | .001 b |
| 2/69 (2.9) a | 14/63 (22.2) a | 11/233 (4.7) a | .50 a , b | .001 a | .001 a , b | |
| Self-reported symptom | ||||||
| Abnormal vaginal discharge | 31/97 (32.0) | 68/119 (57.1) | 108/442 (24.4) | .12 | < .001 | < .001 |
| 31/84 (36.9) a | 66/106 (62.3) a | 103/323 (31.9) a | .38 a | < .001 a | < .001 a | |
| Lower abdominal pain | 58/98 (59.2) | 89/125 (71.2) | 287/442 (65.0) | .28 | .19 | .06 |
| 51/85 (60.0) a | 76/108 (70.4) a | 208/323 (64.4) a | .45 a | .26 a | .13 a | |
| (Post)-coital bleeding | 21/94 (22.3) | 36/126 (28.6) | 52/437 (11.9) | .008 | < .001 | .22 |
| 21/83 (25.3) a | 34/107 (31.8) a | 48/321 (14.9) a | .03 a | < .001 a | .33 a | |
| 19/87 (21.8) d | 34/114 (29.8) d | 47/383 (12.3) d | .02 d | < .001 d | .20 d | |
| Painful urination | 9/97 (9.3) | 23/122 (18.8) | 37/445 (8.3) | .76 | .001 | .05 |
| 9/84 (10.7) a | 22/108 (20.4) a | 32/324 (9.9) a | .82 a | .004 a | .07 a | |
| 9/90 (10.0) d | 19/114 (16.7) d | 28/385 (7.3) d | .38 d | .003 d | .17 d | |
| Prolonged menstrual cycle (>7 d) | 5/96 (5.2) | 3/122 (2.4) | 18/435 (4.1) | .60 | .39 b | .28 b |
| 5/83 (6.0) a | 2/107 (1.9) a | 18/319 (5.6) d | .89 a | .11 a , b | .13 a , b | |
| Signs/symptoms combined e | ||||||
| Cervical tenderness, postcoital bleeding, abnormal vaginal discharge | 21/92 (22.8) | 51/115 (44.3) | 40/409 (9.8) | .01 | < .001 | .001 |
| 17/74 (23.0) c | 45/93 (48.4) c | 28/331 (8.5) c | .001 c | .001 c | .001 c | |
| 21/88 (23.9) d | 49/110 (44.5) d | 37/368 (10.0) d | < .001 d | < .001 d | .002 d |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
