Mycoplasma genitalium Infection in Young Women Without Urogenital Symptoms Presenting to a Community-based Emergency Department in Birmingham, Alabama

Abstract

We used the FDA-cleared Aptima Mycoplasma genitalium assay to evaluate for M. genitalium infection among young women without urogenital symptoms presenting to a community-based emergency department in Birmingham, Alabama between August 2016 to August 2019 for evaluation of non-gynecological concerns. M. genitalium was detected in 23 of 155 (14.8%) women.

Short Summary

Among 155 women without urogenital symptoms who presented to a community-based emergency department in Birmingham, Alabama, for evaluation of non-gynecological concerns, 14.8% tested positive for Mycoplasma genitalium.

INTRODUCTION

Mycoplasma genitalium (MG) infection is an important sexually transmitted infection (STI) associated with urethritis, cervicitis, pelvic inflammatory disease, and perinatal complications.^1,2 MG infection shares similar risk factors with some other urogenital bacterial STIs,^3–6 and the prevalence of MG infection has exceeded 20% in some studies of patients attending STI clinics or sex-workers.^4,7 Unfortunately, treating MG is challenging. Tetracyclines have low efficacy for MG infection (although MG has no reported in vitro resistance to tetracyclines), and macrolide efficacy is declining due to a rise in prevalence of MG strains with macrolide resistance.⁸ Therefore, empiric treatment of MG infection with antibiotics typically used for bacterial STIs may frequently result in treatment failure.

Many MG-infected individuals are asymptomatic. Because the clinical significance and natural history of asymptomatic MG infections are poorly understood, MG screening in asymptomatic individuals is not recommended. The prevalence and risk factors of asymptomatic MG infection in different US populations are also poorly understood, in part due to scarcity of MG infection prevalence studies among asymptomatic individuals. In addition, prior to Food and Drug Administration (FDA) clearance of two MG nucleic acid amplification tests (NAATs) in 2019,⁹ previous studies utilized a wide variety of in-house and commercially-available research-based MG NAATs. While the comparative performance of different MG NAATs are not well-characterized, emerging evidence suggests that the FDA-cleared Aptima® Mycoplasma genitalium (AMG) assay (Hologic Inc., San Diego, CA) may be more sensitive than some other MG NAATs.^10–13 Therefore, interpretation of earlier studies on prevalence of MG infection may be limited by the use of lower sensitivity research-based MG NAATs in some studies.

We conducted a cross-sectional study of MG infection in a convenience sample of young asymptomatic women seen in a community-based emergency department. The purpose of this study was: (1) to determine the proportion of young asymptomatic women in this clinical setting who tested positive for MG using the AMG assay; and (2) describe the demographic, behavioral, and clinical characteristics associated with asymptomatic MG infection.

MATERIALS & METHODS

This study utilized data and urogenital samples collected at enrollment in an ongoing prospective chlamydia study in females aged 16–29 years without urogenital symptoms who present to a community-based emergency department in Birmingham, AL for evaluation of non-gynecological concerns; the study began enrolling in August 2016. Exclusion criteria include current pregnancy, known HIV infection, history of use of oral antibiotics that have anti-chlamydial activity within the last one month, current immunosuppressive therapy, an active immunosuppressive condition (e.g., malignancy, autoimmune disease, or diabetes mellitus), history of hysterectomy, allergy to macrolide antibiotics, and no lifetime history of sexual activity. Participants provide a self-collected vaginal swab for Chlamydia trachomatis (CT) and Neisseria gonorrhoeae (NG) testing (Aptima Combo 2® assay, Hologic Inc., San Diego, CA) per the manufacturer’s protocol and a first-catch urine sample. Participants also complete a standardized interview that collects data on demographics, self-reported lifetime history of STIs and vaginal infections, hormonal contraception use, and sexual behaviors. This study was approved by the University of Alabama at Birmingham Institutional Review Board.

MG testing was conducted using the FDA-cleared AMG assay run on the Panther® system according to the manufacturer’s protocol. Most MG testing (88.4%) was performed on residual vaginal swab specimens stored in Aptima media at −20°C, however the remainder of testing was performed on first-catch urine stored at −20°C (due to some participants having insufficient vaginal specimen remaining for testing). To evaluate differences in demographic, clinical, and behavioral characteristics between women with versus without MG detected, we used Fisher’s exact and Chi-square tests for categorical variables and the Wilcoxon rank sum test for continuous variables. We excluded missing values from analysis and used P < 0.05 as a cutoff for statistical significance.

RESULTS

A total of 155 participants were enrolled in the emergency department between August 29, 2016 to August 1, 2019 (Table 1): the median age was 25 years (range 16–29); 77% were African American and 1.9% were Hispanic. Regarding reported sexual behaviors during the 3 months prior to enrollment: 89% were sexually active, the median sexual partner number was 1 (range of 0–3), and 25.4% reported a new sexual partner. Regarding reported prior urogenital infections, 69.7% reported prior vulvovaginal candidiasis, 34.2% prior CT infection, 28.4% prior bacterial vaginosis, and 23.9% prior trichomoniasis.

Table 1.

Comparison of Characteristics Between Asymptomatic Women Who Tested Positive Versus Negative for Mycoplasma genitalium (MG)

	Total (N = 155)	MG Positive (N = 23)	MG Negative (N = 132)	P-value
Age, median (range)	25 (16 – 29)	24 (19 – 29)	25 (16 – 29)	0.36*
African American, N (%)	119 (76.8%)	23 (100%)	96 (72.7%)	0.002†
Hispanic, N (%)	3 (1.9%)	0 (0.0%)	3 (2.3%)	1.00†
Sexually active in last 3 months, N (%)	138 (89.0%)	23 (100%)	115 (87.1%)	0.08†
Sexual partner number last 3 months, median (range)	1 (0 – 3)	1 (1 – 2)	1 (0 – 3)	0.002*
New partner in last 3 months, N (%)	35 (25.4%)	9 (39.1%)	26 (22.6%)	0.10††
Gender of partner(s) last 3 months, N (%)				0.13†
Male	121 (87.7%)	20 (87.0%)	101 (87.8%)
Female	13 (9.4%)	1 (4.4%)	12 (10.4%)
Both male and female	4 (2.9%)	2 (8.7%)	2 (1.7%)
Number of days since last sex with any partner, median (range)	4 (0 – 90)	3 (1 – 90)	5 (0 – 90)	0.04*
Hormonal contraceptive use, N (%)	39 (25.2%)	4 (17.4%)	35 (26.5%)	0.44†
Self-reported lifetime history
Bacterial vaginosis	44 (28.4%)	7 (30.4%)	37 (28.0%)	0.81††
Candidiasis	108 (69.7%)	16 (69.6%)	92 (69.7%)	0.99††
Chlamydia	53 (34.2%)	10 (43.5%)	43 (32.6%)	0.31††
Gonorrhea	21 (13.5%)	7 (30.4%)	14 (10.6%)	0.01††
Trichomoniasis	37 (23.9%)	5 (21.7%)	32 (24.2%)	1.00†
Current infections
Positive C. trachomatis NAAT, N (%)	9 (5.8%)	3 (13.0%)	6 (4.6%)	0.13†
Positive N. gonorrhoeae NAAT, N (%)	2 (1.3%)	0 (0.0%)	2 (1.5%)	1.00†

^*= Wilcoxon rank sum test;

^†= Fisher’s exact test;

^††= Chi-square test of independence

N=1 missing sexual partner number last 3 months (in MG positive group)

N=17 reporting no sexual activity in the last 3 months (all in MG negative group), resulting in N = 17 missing for the following variables: new partner in last 3 months, number of days since last sex with any partner, gender of partner(s) last 3 months

NAAT, nucleic acid amplification test

CT, NG, and MG were detected in 9 (5.8%), 2 (1.3%), and 23 (14.8%) participants, respectively. In total, 13.0% of MG positive participants were coinfected with CT, while 0% were coinfected with NG. Of the aforementioned demographic, clinical, and behavioral characteristics, MG detection was significantly associated with African American race (P = 0.002), fewer number of days since last sexual activity (P = 0.04), higher number of sexual partners in the last 3 months (P = 0.002), and self-reported history of NG infection (P = 0.02) (Table 1). There was a trend towards a higher MG detection frequency in those reporting being sexually active in the last 3 months and having a new sexual partner during that time (P ≤ 0.10).

DISCUSSION

Our study is one of only a few US-based studies to evaluate for MG infection in asymptomatic women using the FDA-cleared AMG assay, which has improved sensitivity compared to some other research-based assays.^10–13 We found that in women presenting to a community-based emergency department without urogenital symptoms, 14.8% tested positive for MG. Previous US studies varied widely in their estimates of MG infection prevalence. Those enrolling patients complaining of urogenital symptoms or that recruited directly from STI clinics detected MG in 7.0–22.4% of patients.^4,14–17 In contrast, a couple of US studies which primarily or exclusively enrolled asymptomatic individuals at sites other than STI clinics had lower estimates of MG infection prevalence.^5,18 In the only US population-based assessment of MG infection prevalence, the estimated prevalence of MG infection among individuals aged 18–27, most who were asymptomatic, was 1.0%.⁵ Likewise, in a cross-sectional study of men enrolled from an urban emergency department in Seattle who denied urogenital symptoms, MG was detected in only 3.0%.¹⁸ Thus, a source of variation in estimates of asymptomatic MG infection prevalence within the US is the heterogeneity between study designs, study populations, and lack of more population-based studies using probability-sampling methods. This may partially explain why our study, conducted in an urban community-based emergency department, had a higher percentage of asymptomatic patients who tested positive for MG compared to the aforementioned studies of asymptomatic MG infection.^5,18

Previous reports of MG infection prevalence have also been limited by the lack of a gold-standard test. Because MG culture is prohibitively difficult, studies have relied on NAAT for MG detection. Until 2019, all NAATs used in previous US-based studies were research-based assays. For many of these research-based NAATs, little is known regarding their comparative performance characteristics, in particular their sensitivities. This is important, as MG infection prevalence estimates could differ considerably with assays that vary widely in sensitivity. Indeed, recent reports comparing MG NAATs suggest the FDA-cleared AMG assay has a higher sensitivity than some other MG NAATs.^10–13 Given that the two US-based studies of asymptomatic MG infection mentioned above both used research-based PCR assays,^5,18 it is possible that the higher frequency of asymptomatic MG infection in our study may be partly attributable to the higher sensitivity of the AMG assay; however, direct comparisons of the AMG assay to the NAATs used in those studies is needed to confirm this.

We found significant associations between MG detection and select demographic and behavioral factors. African American race was strongly associated with MG detection, consistent with previous reports.^3–5 MG detection was significantly associated with an increased sexual partner number in the previous 3 months and a shorter interval since the most recent sexual activity. MG infection often spontaneous clears, although the precise timing and factors affecting clearance remain to be elucidated.^{4, 19–21} The association of MG detection with more recent sexual activity may reflect a tendency to detect MG more frequently among women with recent sexual encounters, whereas women with a longer time interval between last sexual intercourse and MG testing may have more time to spontaneously clear their infection.

Our study is not without limitations. First, because our sample size was limited, we may not have had adequate power to detect more modest associations between MG infection and participant demographic and behavioral characteristics. Second, our study population consisted of women evaluated at an urban emergency department in Birmingham, AL. In US clinical populations, patients presenting to the emergency department tend to be of lower socioeconomic status and medically underserved relative to other clinical populations.²² Thus, findings from our study population may be not be generalizable to the general population in Birmingham or the U.S. Moreover, given that 77% of our patients were African American and that African American race has been associated with higher MG infection prevalence,^3–5 the proportion of asymptomatic women with MG detected that is reported here may not reflect MG detection rates in asymptomatic populations with a different distribution of race/ethnicity.

In summary, we demonstrated that among young women without urogenital symptoms presenting to a community-based emergency department for evaluation of non-gynecological concerns, MG infection was detected in 14.8% of women, a frequency much higher than that reported in other studies that included mostly or all asymptomatic individuals. This may be explained by the geographic differences in the rates of MG infection, the demographic and socioeconomic characteristics of the population we studied, and our use of a more sensitive MG assay than earlier studies. In order to better understand the public health implications of asymptomatic MG infection and to guide screening and treatment recommendations, more population-based studies using probability samples and high-sensitivity tests, such as the AMG assay, are needed to establish generalizable estimates of MG infection prevalence and risk factors. Moreover, because the sequelae of asymptomatic MG infection are largely unknown, studies assessing reproductive outcomes in women with asymptomatic MG infection are needed.