The association between Mycoplasma genitalium and pelvic inflammatory disease after termination of pregnancy
Dr C Bjartling, Institute of Clinical Sciences, Department of Obstetrics and Gynaecology, Malmo University Hospital, SE-205 01 Malmo, Sweden. Email email@example.com
Please cite this paper as: Bjartling C, Osser S, Persson K. The association between Mycoplasma genitalium and pelvic inflammatory disease after termination of pregnancy. BJOG 2010;117:361–364.
The prevalence and complications of Mycoplasma genitalium and Chlamydia trachomatis infections among women undergoing termination of pregnancy were studied in this nested case–control study at Malmo University Hospital, Sweden, during 2003 to 2007. The study comprised 2079 women presenting for termination of pregnancy. Forty-nine women with M. genitalium infection and 51 women with C. trachomatis infection, together with 168 negative control women, were evaluated. The prevalences of M. genitalium and C. trachomatis were 2.5% and 2.8%, respectively. The M. genitalium was strongly associated with post-termination pelvic inflammatory disease (odds ratio 6.29, 95% CI 1.56–25.2). The increased risk for pelvic inflammatory disease associated with M. genitalium infection after termination of pregnancy suggests a causal relationship.
Provide feedback or get help You are viewing our new enhanced HTML article.
If you can't find a tool you're looking for, please click the link at the top of the page to "Go to old article view". Alternatively, view our Knowledge Base articles for additional help. Your feedback is important to us, so please let us know if you have comments or ideas for improvement.
Mycoplasma genitalium has been associated with nongonococcal urethritis in men1 and is also recognised as a sexually transmitted agent in men and women. In women it has been associated with cervicitis in several studies, although some reports have failed to detect such a link.2 The role of M. genitalium in pelvic inflammatory disease (PID) is still uncertain although the organism has been demonstrated in both endometrial biopsies and in a single sample from the fallopian tubes.3 Antibodies to M. genitalium have been shown to be an independent risk factor for tubal factor infertility.4
Chlamydia trachomatis is a well-recognised cause of PID and has repeatedly been shown to be a risk factor for post-termination endometritis and salpingitis. Mycoplasma genitalium has been reported in 8.7% of women below 25 years of age seeking termination of pregnancy in New Zealand and in only 0.5% in Denmark.5,6 Post-termination complications were not assessed in any of these studies.
In this study women undergoing termination of pregnancy were examined. Women who tested positive for M. genitalium and C. trachomatis, together with women who tested negative, used as controls, were compared with regard to post-termination complications such as PID.
The study was performed in the outpatient service at the Department of Gynaecology at Malmö University Hospital in Sweden. Women seeking termination of pregnancy (TOP) are routinely screened for C. trachomatis. These women were invited to participate in a study on M. genitalium. The sample obtained for C. trachomatis testing was also used for M. genitalium testing and the results for M. genitalium were disclosed only after the study was terminated. The study started in November 2003 and continued until June 2007 with two interruptions because of limited testing capacity.
Age-matched women without M. genitalium or C. trachomatis infection were chosen as controls from among women tested during the same month as the positive index case. Control women were also matched for the procedure used (medical or surgical) because post-termination infection has been reported to be higher following surgical TOP. All the women in the control group were used as controls for both C. trachomatis-infected women and M. genitalium-infected women. Age may therefore be a confounder which was adjusted for in the statistical analyses.
According to clinical routine at the gynaecological department, medical TOP was induced with 200 mg mifepristone orally on day 1 followed by 0.2 mg misoprostol orally on day 3.
Surgical TOP was performed by vacuum aspiration under general anaesthesia. All women who were C. trachomatis-positive were treated with doxycycline, 200 mg orally day 1 and 100 mg daily for 9 days, starting at least 5 days before TOP.
After the study was closed the medical records of women infected with M. genitalium, women infected with C. trachomatis and of women who were negative controls were collected. A standardised form was used containing a short medical history and clinical assessment. The diagnosis of post-termination infection according to WHO international classification of diagnoses (endometritis/salpingitis, O04.1) was compared between women in the infected and control groups. The clinical diagnosis of post-termination infection was based on the same criteria as for PID using the guidelines from Centers for Disease Control and Prevention (Atlanta, GA, USA). Lower abdominal pain, cervical, uterine or adnexal tenderness at pelvic bimanual examination together with one of the following signs: pathological vaginal wet smear or yellow pus from the endocervical canal, elevated C-reactive protein >8 or fever (oral temperature >38.0°C). Infections occurring within 4 weeks after the TOP were considered as post-termination infections. The follow up after TOP was 6 weeks by tracing the diagnosis in medical records. All women undergoing TOP were urged to contact the clinic after the procedure if they had any symptoms of infection. This information was given also in a written form.
Urine together with cervical samples or (from 2005) urine together with vaginal samples were obtained for C. trachomatis testing. During the study period two different tests were used for C. trachomatis. From the start of the study until mid-2006 the Cobas Amplicor (Roche, Basel, Switzerland) was used. After that, the m2000™ (Abbott, Abbott Park, IL, USA) was used. In 2006 a new variant of C. trachomatis was discovered in Sweden with a deletion in the cryptic plasmid that contained the target sequences for both these tests. The new variant of C. trachomatis constituted 25–30% of the infected women in 2006 and must have appeared after 2000/1 when no such strains could be detected among 258 culture-positive women tested in retrospect.
From 2007 the m2000 test had been modified to detect the new variant as well. Mycoplasma genitalium was detected using the polymerase chain reaction (PCR) and targeting the operon of one of the surface adhesion proteins, MgPa, as described by others.7 The primers MgPA-1: 5′-AGTTGATGAAACCTTAACCCCTTGG were used together with MgPa-3: 5′-CCGTTGAGGGGTTTTCCATTTTTGC. Amplicons were identified after gel electrophoresis and staining by ethidium bromide. Ambiguous bands were sometimes observed. To increase specificity, a semi-nested PCR was finally used where the MgPa-1 primer was combined with the MgPa-2 5′-GACCATCAAGGTATTTCTCAACAGC. Strains detected in this way were later DNA sequenced to verify their identity. Samples obtained for C. trachomatis testing were used to test for M. genitalium as well. Stored samples were pooled with five to ten samples in each pool and then submitted to PCR. Positive pools were then resolved by testing the individual samples of the pool separately.
Data were analysed using SPSS for Windows version 17.0 (IBM, Chicago, IL, USA). Frequencies of gynaecological characteristics, symptoms, clinical signs and diagnoses (categorical variables) were compared between women who were M. genitalium-positive, women who were C. trachomatis-positive and the women who were M. genitalium- and C. trachomatis-negative controls using the Pearson chi-square test or the Fishers’ exact test. Backwards stepwise binary logistic regression was used to adjust for possible confounders such as age, M. genitalium and C. trachomatis. The 95% confidence intervals were calculated.
In all, 2079 of 2156 (96.4%) women requesting TOP agreed to participate and were tested for both M. genitalium and C. trachomatis infection. Mycoplasma genitalium was detected in 52 of 2079 (2.5%) and C. trachomatis was detected in 59 of 2079 (2.8%) of these women. Forty-nine M. genitalium-infected women were included because three women did not proceed with TOP. Fifty-one C. trachomatis-infected women were included because seven medical records could not be retrieved and one woman did not proceed with TOP. In all, 168 women were included as negative controls. Three patients were co-infected with both M. genitalium and C. trachomatis. They were excluded from the Pearson chi-square test.
The mean age of the women who were M. genitalium-positive was 25.6 years (range 17–40 years) and the mean age of C. trachomatis-positive women was 22.5 years (15–36 years). The highest frequencies of both C. trachomatis and M. genitalium were seen in the youngest women. The mean age of the women who acted as negative controls was 23.6 years (17–40 years; data not shown).
Six of forty-nine (12.2%) women with M. genitalium infection had complications compatible with PID and were diagnosed with post-termination infection whereas only four of 168 (2.4%) among the women who were M. genitalium- and C. trachomatis-negative controls were diagnosed with post-termination infection. (Table 1).
Table 1. Post-termination pelvic inflammatory disease (PID) and association with Mycoplasma genitalium in 217 women undergoing termination of pregnancy; logistic regression was used
|Medical and surgical termination||6/49 (12.2)||4/168 (2.4)||6.29 (1.56–25.2)||0.01|
|Surgical termination||3/16 (18.8)||3/70 (4.3)||5.78 (1.02–32.80)||0.05|
|Medical termination||3/33 (9.1)||1/98 (1.0)||8.68 (0.58–14.84)||0.12|
Among the 49 women who were M. genitalium-positive, three had concurrent C. trachomatis infection. One of them was diagnosed with post-termination PID. When excluding them from the analyses there were five women with PID among the 46 women in the M. genitalium-infected group compared with four of the 168 women in the negative control group. This difference was statistically significant (P = 0.02) using Fisher’s exact test (data not shown). This difference remained statistically significant after adjustment for age and concomitant C. trachomatis infection using logistic regression analysis (odds ratio [OR] 6.29, 95% CI 1.56–25.2) (Table 1).
Stratifying the women according to procedure into medical and surgical TOP, M. genitalium remained an independent risk factor for PID in surgical TOP (OR 5.78, 95% CI 1.02–32.80) but not in medical TOP (OR 8.68, 95% CI 0.58–14.84) (Table 1).
All the 51 women infected with C. trachomatis that were included in this study were treated with doxycycline before TOP. Among these women, no single post-termination PID was observed.
One patient had a double infection by both M. genitalium and C. trachomatis. She underwent medical termination and developed post-termination PID after receiving 10 days of treatment with doxycycline starting 5 days before TOP. This patient was excluded from the statistical analysis.
Neisseria gonorrhoeae is very rare in Sweden with 7.85 reported cases/100 000 inhabitants in 2008. Testing for N. gonorrhoeae is not included in the opportunistic screening at the gynaecological outpatient service in Malmö but testing is performed in selected cases where the medical history, symptoms or clinical signs arouse suspicion of a possible infection. During the whole study period, 1390 patients visiting the outpatient service were tested for N. gonorrhoeae. No positive case was detected.
Mycoplasma genitalium infection was detected in 2.5% and C. trachomatis in 2.8% in this group of women seeking TOP. The highest frequency of M. genitalium infections was found in women aged between 20 and 35 years, gradually declining with age. The age distribution was similar for C. trachomatis but the age-related decline was more pronounced. These patterns are expected for sexually transmitted infections affecting the general population.
Two previous studies have reported frequencies of M. genitalium in women seeking TOP. One such study from Denmark found only 0.5%M. genitalium-infected women and another study from New Zealand found 8.7% among women younger than 25 years of age. Complication rates were not reported in those two studies. In this study the prevalence of M. genitalium was 2.5%.
It has previously been reported that C. trachomatis is a risk factor for post-termination PID with a frequency of infection of more than 20% in women who were C. trachomatis-positive.8 Other studies have shown similar results. In the present study all patients seeking TOP were routinely screened for C. trachomatis infection and were treated with doxycycline before TOP. No case of post-termination PID was observed among these patients. Treatment therefore seemed to be highly effective.
One woman, who developed post-termination PID, was infected by both C. trachomatis and M. genitalium and had received treatment with doxycycline before termination. Although treatment with doxycycline is generally considered effective for C. trachomatis, it fails to eradicate M. genitalium in a high proportion of cases. The post-termination PID in this case was therefore probably related to M. genitalium infection but the case was excluded from statistical analysis as it could not be properly classified.
Post-termination PID was detected in 12.2% of the women with M. genitalium infection compared with 2.5% among the women who acted as controls. This difference was statistically significant even after adjustment for age and concomitant C. trachomatis infection. This is the first time to our knowledge that such an association has been reported.
Given a prevalence of M. genitalium infection of 2.5% among these women this infection accounted for 0.3% of all cases of post-termination PID, which is about one-tenth of the risk observed among those negative for both M. genitalium and C. trachomatis.
Pelvic inflammatory disease after TOP associated with M. genitalium could probably be prevented if routine screening for M. genitalium were performed. The same sample could be used for both C. trachomatis and M. genitalium testing in most settings. Azithromycin or moxifloxacin are effective both against M. genitalium and C. trachomatis. Whether such dual testing should be considered in women undergoing TOP partly depends on the risk of tubal damage after M. genitalium infection, which is not fully known. It also depends on the prevalence in the population seeking TOP, which seems to differ highly between settings. Confirmation of the results presented in this study is also urgently needed.
There are some limitations of the present study. Testing for N. gonorrhoeae was not performed routinely. However, no positive case was found among 1390 tested patients seeking care at the outpatient service. N. gonorrhoeae therefore seemed unlikely to be of importance in this study.
Mycoplasma genitalium was found to be associated with PID after termination of pregnancy. Testing for M. genitalium and treatment before TOP to prevent such infections seems feasible. Screening for M. genitalium in this group of patients might be considered but before it could be generally recommended confirmatory studies are needed.
Disclosure of interest
We declare that we have no conflict of interest.
Contribution to authorship
All authors had the idea for and designed the study, took part in the interpretation of results, and approved the final version.
Details of ethics approval
This study was part of a more comprehensive study of M. genitalium manifestations in women at a gynaecological department and had been approved by the local ethics committee. Informed consent was obtained from the patients included in the study.
No special funding was received. This study was partly supported by Lund University, Faculty of Medicine, Sweden.