Abstract
Objectives:
Bacterial vaginosis-associated bacterium 2 (BVAB2), Mageeibacillus indolicus, and Sneathia spp. are highly predictive of bacterial vaginosis (BV) in cisgender women. They have been associated with nongonococcal urethritis (NGU) in cisgender men in some but not all populations. We evaluated this association in a cross-sectional study of cis-men who have sex with women only (MSW).
Methods:
MSW without gonorrhea attending a sexual health clinic from 2014-2018 completed a computer-assisted self-interview, clinical interview, and examination. NGU was defined as ≥5 polymorphonuclear leukocytes/hpf in urethral exudates plus either urethral symptoms or visible discharge. Urine was tested for Chlamydia trachomatis and Mycoplasma genitalium using Aptima (Hologic, Inc.) and for BVAB2, Mageeibacillus indolicus, Sneathia spp., Trichomonas vaginalis, Ureaplasma urealyticum, Haemophilus influenzae, herpes simplex virus, and adenovirus using quantitative PCR.
Results:
Of 317 MSW ages 17-71, 67 (21.1%) had Sneathia spp., 36 (11.4%) had BVAB2, and 17 (5.4%) had M. indolicus at enrollment. Having ≥3 partners in the past two months was the only characteristic that was more common among MSW with than without these bacteria (BVAB2: 47% vs. 23%; M. indolicus: 53% vs. 24%; Sneathia spp.: 42% vs. 22%, p≤0.03 for all). One-hundred seventeen men (37%) were diagnosed with NGU at enrollment. There was no significant association of BVAB2, M. indolicus, or Sneathia spp. with NGU (aOR=0.59 [0.14-2.43]; aOR=3.40 [0.68-17.06]; aOR=0.46 [0.16-1.27]). Of 109 MSW with monthly samples, 34 (31.2%) had one of the bacteria at one or more follow-up visits, 22 of which were co-colonized with >1. Median persistence over 6 months did not differ significantly (BVAB2=30.5 days [IQR=28-87]; M. indolicus=87 days [IQR=60-126]; Sneathia spp.=70 days [IQR=30-135]; p≥0.20 for each comparison).
Conclusions:
Neither BVAB2, M. indolicus, nor Sneathia spp. were associated with increased risk of prevalent NGU in MSW attending a sexual health clinic.
Keywords: Non-gonococcal urethritis, Sneathia, BVAB2, Mageeibacillus indolicus, bacterial vaginosis, men who have sex with women (MSW)
Short Summary:
Among cisgender men who have sex with women, three bacterial-vaginosis associated bacteria were associated with higher number of female partners but not associated with increased likelihood of non-gonococcal urethritis.
INTRODUCTION
Urethritis is the most common male urethral syndrome in the United States. Gonococcal urethritis is often less common than nongonococcal urethritis (NGU), which is typically caused by Chlamydia trachomatis (CT) or Mycoplasma genitalium (MG)1. Trichomonas vaginalis (TV), herpes simplex virus (HSV) and adenovirus also cause some cases2 and Haemophilus influenzae and Mycoplasma penetrans have recently been associated with NGU3,4. Ureaplasma urealyticum (UU) has been associated with NGU in some but not all studies, and only in those that used quantitative assays3,5,6. Bacteria associated with bacterial vaginosis (BV) have been explored as potential causes of NGU2,7,8 with mixed results.
BV is a common female reproductive tract syndrome9 and several individual BV-associated bacteria (BVAB) have been highly predictive of and specific for BV10. Some have also been associated with cervicitis, including Mageeibacillus indolicus (previously BVAB3), Sneathia spp. and BVAB211. These BVAB may be sexually transmitted12,13 and may result in similar inflammatory conditions in the male urogenital tract.
We previously conducted a case-control study among Seattle men and found that Sneathia spp. detected by species-specific quantitative PCR (qPCR) were significantly associated with idiopathic NGU. BVAB2 and M. indolicus were more common among men with than without NGU2. In contrast, a case-control study among Swedish men using broad-range 16S rRNA gene PCR and sequencing found no association between Sneathia spp. and idiopathic NGU[7]. A separate cross-sectional study using more sensitive species-specific qPCR also detected no association between either Sneathia sanguinegens or Sneathia amnii and NGU8.
Given the inconsistencies between studies, we sought to evaluate the association between NGU and BVAB2, M. indolicus, and Sneathia spp. in a population of cisgender men who have sex with women only (MSW). We also sought to estimate the prevalence of and characteristics associated with these three BVAB and describe the natural history of urethral colonization with them in a subset of MSW with longitudinal data.
METHODS
Study Design and Procedures
Cisgender MSW attending the Public Health–Seattle & King County (PHSKC) Sexual Health Clinic (SHC) in Seattle, Washington from 2014-2018 enrolled in a longitudinal study of the male urethral microbiota. Detailed study procedures have been previously described14. Briefly, persons assigned male sex at birth attending the SHC who were 16 years or older, who reported sexual activity during the past 60 days, and who had exclusively female partners in the past year were eligible. MSW who reported taking any antibiotics in the past 30 days, had evidence of gonorrhea on Gram stain, or had contact to a partner with gonorrhea were excluded. After providing informed consent, participants underwent a clinical interview and examination. Both symptomatic and asymptomatic men provided urethral swab and first void urine specimens. Polymorphonuclear leukocytes (PMNs) were quantified in a Gram-stained urethral smear. All clinic attendees completed a clinic computer-assisted self-interview (CASI) providing basic socio-behavioral information. Enrolled MSW reported more detailed behavioral data via a study CASI. Data on urethral symptoms, examination findings and medical history were abstracted from paper medical charts.
NGU was defined as ≥5 PMNs per high power field (hpf) plus either urethral symptoms or visible discharge in the absence of gonorrhea. Men with NGU were treated according to clinic standard of care, typically azithromycin (1g single dose) or doxycycline (100 mg twice daily x 7 days). Enrolled men attended study visits once monthly for six months. Follow-up study procedures were identical to those at enrollment.
Urine specimens were tested for Neisseria gonorrhoeae and CT using Aptima Combo2 while Mycoplasma genitalium (MG) was detected using Aptima with analyte-specific reagents (Hologic, Inc. San Diego, CA). A subset of 174 men who participated in a separate case-control study were tested for Trichomonas vaginalis (TV), Urea plasma urealyticum (biovar 2 or UU), Haemophilus influenzae (HI), herpes simplex virus (HSV)15, and adenovirus16 using qPCR3. We tested the first 500 urine specimens (322 enrollment; 178 follow-up specimens) for BVAB2, M. indolicus, and Sneathia spp. using species-specific qPCR17 to evaluate their association with NGU.
Statistical Analysis
We compared characteristics of participants with and without NGU at enrollment, and those with and without each BVAB using Fisher’s exact tests, two-sample t-tests with unequal variances, and Wilcoxon rank-sum tests. Examined characteristics included sociodemographic characteristics (age, race/ethnicity, education), sexual behaviors (condom use, number of recent and new sex partners), clinical characteristics (urethral symptoms, presence of urethral discharge, prior STI and NGU diagnoses) and organisms associated with NGU.
We utilized multivariable logistic regression to evaluate the association between presence of each BVAB and NGU at enrollment, determining a priori to adjust for CT, MG, UU, HI, the other BVABs, and characteristics significantly associated with any BVAB in bivariate analyses at p<0.10. We also tested interaction terms between each BVAB in bivariate models (BVAB2*Sneathia spp; BVAB2*M. indolicus; Sneathia spp*M. indolicus). As only a subset of men were tested for both UU (n=174) and HI (n=182), we included a category for missing test results to retain them in the analysis. Given that this subset constituted only 55% of our study population, we conducted a sensitivity analysis, restricting the model to individuals with test results for UU and HI (Supplementary Table S1). We also used logistic regression to evaluate the association between the quantity in log copies / mL of each BVAB and NGU at enrollment, adjusting for the same variables and conducted a similar sensitivity analysis restricted to persons with complete test results (Supplementary Table S2). Individuals without BVAB were assigned a quantity of half the threshold value in the sample.
Additionally, we conducted three sub-analyses. In the first, we assessed the relationship between these bacteria and idiopathic NGU at enrollment by restricting analyses to those without CT, MG, and known infections with UU, TV, HSV, and adenovirus to be consistent with our prior study2. In the second, we evaluated the influence of how NGU was defined, comparing associations of NGU with these BVAB using our study definition to associations using two alternative definitions. The “Objective Signs-Based” definition was employed in our previous study2 and defined NGU as clinician-observed discharge OR ≥5 PMNs/hpf. The “Symptoms-Based” definition defined NGU as urethral symptoms PLUS either clinician observed discharge or ≥5 PMNs/hpf. In the third sub-analysis, we described the natural history of infection up to 6 months in the subset of 109 men with follow-up specimens. We constructed a heatmap summarizing the presence and quantity of each BVAB, NGU status, signs, symptoms, MG and CT infections, and antibiotic use. BVAB were considered persistent if they were detected at ≥2 consecutive visits. Duration was calculated as the number of consecutive days for which the BVAB was detected over at least two consecutive visits. In a sensitivity analysis, we reclassified a negative test bracketed by positive tests as positive to assess the influence of possible false negative results on estimated duration.
For these analyses we used Stata V.16 and two-sided tests with a significance-level of 0.05. The parent study procedures were approved by the University of Washington Institutional Review Board. These analyses were restricted to anonymized data and required no additional approvals.
RESULTS
Of 322 MSW with BVAB qPCR results, five were excluded post-hoc as ineligible, leaving 317 MSW in these analyses. Of these, 199 were included in an earlier report evaluating NGU in MSW and in men who have sex with men (MSM)18 and 185 were included in an earlier report evaluating the urethral microbiota and NGU3 (Supplementary Figure S1). The average age was 33.8 years (SD ± 10.2 years; range=17-71 years). Most (46.1%) were non-Hispanic White or non-Hispanic Black (22.4%) persons, while 36 reported Hispanic/Latino ethnicity (11.4%) and 64 reported other/multiple races or did not report their race (20.2%). One hundred seventeen MSW had NGU at enrollment (36.9%). Seventy-one (22.4%) had a BVAB detected, of whom 35 (50.7%) had >1. Thirty-six participants (11.4%) had BVAB2, 17 (5.4%) had M. indolicus, and 67 (21.1%) had Sneathia spp. detected by qPCR. The mean quantities were 2.65 (±0.93), 2.82 (±1.22), and 3.81 (±1.34) log copies / mL for BVAB2, M. indolicus, and Sneathia spp., respectively.
Characteristics associated with prevalent NGU
As previously reported3,14,18, MSW who were non-Hispanic Black, had <high school education, reported inconsistent condom use, had a history of CT, or who had MG, CT or HI at enrollment were significantly more likely to have NGU than those without these characteristics (p<0.05 for all, data not shown again here). In contrast, participants with and without NGU did not differ significantly by age; sexual history in the past two months; detection of UU, TV, HSV, or adenovirus; or history of NGU or gonorrhea 3,14. Although the 143 MSW missing data for UU, HI, TV, HSV, and adenovirus were significantly less likely to have NGU at enrollment than those with complete data (24% vs. 48%, p<0.001), no other characteristics differed significantly between MSW with and without complete test results (Supplementary Table S3).
Sociobehavioral characteristics associated with BVAB
Having ≥3 sexual partners in the past two months was significantly more common among men with than without BVAB2 (47.2% vs 23.0%, p=0.01), M. indolicus (52.9% vs. 24.2%, p=0.03) and Sneathia spp. (42.2% vs. 21.5%, p<0.01; Table 1). Race/ethnicity was not associated with BVAB2 or Sneathia spp. Men with M. indolicus were somewhat more likely to be non-Hispanic Black (35.3% vs. 21.7%) or Hispanic (23.5% vs. 10.7%) than men without M. indolicus but this was not statistically significant (p=0.10). Men with M. indolicus were also somewhat less likely to report vaginal sex in the past two months than men without M. indolicus (87.5% vs. 97.6%, p=0.075), but this relationship was not statistically significant, nor was it observed with BVAB2 and Sneathia spp. There were no significant differences in age, race, education, consistent condom use, or having a new partner between men with and without BVAB2, M. indolicus, and Sneathia spp.
Table 1 –
Sociobehavioral and Clinical Characteristics by BVAB2, M. indolicus, and Sneathia spp. Status Among Cisgender Men who have Sex with Women Attending the PHSKC Sexual Health Clinic (N=317)
| BVAB2+ (n=36) n (%) |
BVAB2− (n=281) n (%) |
M. indolicus+
(n=17) n (%) |
M. indolicus−
(n=300) n (%) |
Sneathia+
(n=67) n (%) |
Sneathia−
(n=250) n (%) |
|
|---|---|---|---|---|---|---|
| Age, mean (±SD) years | 34.25 (±10.46) | 33.70 (±10.19) | 34.47 (±10.43) | 33.72 (±10.21) | 33.49 (±10.12) | 33.83 (±10.25) |
| Race/Ethnicity | ||||||
| NH White | 14 (38.89) | 132 (46.98) | 4 (23.53) | 142 (47.33) * | 26 (38.81) | 120 (48.00) |
| NH Black | 7 (19.44) | 64 (22.78) | 6 (35.29) | 65 (21.67) | 17 (25.37) | 54 (21.60) |
| Hispanic | 8 (22.22) | 28 (9.96) | 4 (23.53) | 32 (10.67) | 8 (11.94) | 28 (11.20) |
| Othera | 7 (19.44) | 57 (20.28) | 3 (17.65) | 61 (20.33) | 16 (23.88) | 48 (19.20) |
| ≤ HS Education or GEDb | 14 (41.18) | 97 (35.02) | 7 (43.75) | 104 (35.25) | 26 (40.00) | 85 (34.55) |
| Consistent condom useb | 9 (25.00) | 59 (21.77) | 5 (29.41) | 63 (21.72) | 13 (20.31) | 55 (22.63) |
| Sexual History, Past Two Monthsb | ||||||
| Number sex partners | ||||||
| 1 | 10 (27.78) | 117 (42.70) ** | 3 (17.65) | 124 (42.32) ** | 22 (34.38) | 105 (42.68) ** |
| 2 | 9 (25.00) | 94 (34.31) | 5 (29.41) | 98 (33.45) | 15 (23.44) | 88 (35.77) |
| ≥3 | 17 (47.22) | 63 (22.99) | 9 (52.94) | 71 (24.23) | 27 (42.19) | 53 (21.54) |
| ≥1 new sex partners | 28 (80.00) | 193 (69.42) | 14 (82.35) | 207 (69.93) | 44 (66.67) | 177 (71.66) |
| Vaginal sex with SP | 32 (94.12) | 264 (97.42) | 14 (87.50) | 282 (97.58) * | 59 (96.72) | 237 (97.13) |
| Anal sex with SP | 5 (13.89) | 50 (18.18) | 3 (17.65) | 52 (17.69) | 12 (18.46) | 43 (17.48) |
| Oral sex with SP | 30 (96.77) | 248 (91.85) | 13 (92.86) | 265 (92.33) | 57 (95.00) | 221 (91.70) |
| Organisms Detected | ||||||
| C. trachomatis | 1 (2.78) | 34 (12.10) | 1 (5.88) | 34 (11.33) | 3 (4.48) | 32 (12.8) * |
| M. genitalium | 4 (11.11) | 31 (11.03) | 3 (17.65) | 32 (10.67) | 8 (11.94) | 27 (10.80) |
| H. influenzaec (n=182) | 0/19 (0.00) | 14/163 (8.59) | 0/6 (0.00) | 14/176 (7.95) | 0/37 (0.00) | 14/145 (9.66) * |
| HSVd (n=175) | 1/19 (5.26) | 1/156 (0.64) | 0/6 (0.00) | 2/169 (1.18) | 1/37 (2.70) | 1/138 (0.72) |
| Adenovirusd (n=175) | 1/19 (5.26) | 0/156 (0.00) | 0/6 (0.00) | 1/169 (0.59) | 1/37 (2.70) | 0/138 (0.00) |
| T. vaginalise (n=174) | 0/19 (0.00) | 11/155 (7.10) | 1/6 (16.67) | 10/168 (5.95) | 3/37 (8.11) | 8/137 (5.84) |
| U. urealyticume(n=174) | 13/19 (68.42) | 33/155 (21.29) ** | 4/6 (66.67) | 42/168 (25.00) ** | 20/37 (54.05) | 26/137 (18.98) ** |
| History of STD | ||||||
| History of NGUb | 3 (8.57) | 34 (12.14) | 3 (17.65) | 34 (11.41) | 10 (15.15) | 27 (10.84) |
| History of GCb | 4 (11.43) | 40 (14.29) | 4 (23.53) | 40 (13.42) | 8 (12.12) | 36 (14.46) |
| History of CTb | 9 (25.71) | 80 (28.57) | 8 (47.06) | 81 (27.18) * | 21 (31.82) | 68 (27.31) |
% represent percentages excluding missing values
P-values are based on Fisher’s exact tests, while continuous variables were compared using t-tests assuming unequal variance.
Abbreviations: GED = General Educational Development Test, NH = non-Hispanic, PHSKC = Public Health – Seattle & King County, SD = standard deviation, SP = sex partner(s), CT = Chlamydia trachomatis, GC = Neisseria gonorrhoeae, HSV = Herpes simplex virus, NGU = nongonococcal urethritis
Includes other, multiple or unknown race/ethnicity
Missing n =6 for education, n=10 for condom use, n=7 for 2 month sex history, n=4 for 2 month new partners, n=6 for anal sex 2 mo., n=12 for vaginal sex 2 mo., n= 16 for oral sex 2 mo., n=2 for report of GC hx, n=2 for CT hx, n=2 for NGU hx
Only 182 of the 317 participants were tested for HI at enrollment.
Only 175 of the 317 participants were tested for adenovirus and HSV at enrollment.
Only 174 of the 317 participants were tested for UU and TV infection
= <0.1
= <0.05
Clinical characteristics associated with BVAB
Neither BVAB2, M. indolicus, nor Sneathia spp. were significantly associated with detection of MG, HI, TV, HSV, or adenovirus at enrollment, or with reported history of NGU, or gonorrhea (Table 1). Having CT at enrollment was somewhat less common among men with than without each of these three BVAB; history of CT infection was also somewhat more common in men with than without M. indolicus, but none of these relationships were statistically significant. Among those who were tested for UU, detection of UU was significantly more common among men with than without BVAB2 (68.4% vs. 21.3%, p<0.001), M. indolicus (66.7% vs. 25.0%, p=0.04) and Sneathia spp. (54.1% vs. 19%, p<0.001).
BVAB and Prevalent NGU
Neither BVAB2 nor M. indolicus were significantly associated with prevalent NGU (Table 2). In bivariate analyses, men with Sneathia spp. were less likely to have NGU than those without Sneathia spp. (25.4% vs. 40.0%, OR=0.51, 95% CI=0.28-0.93), but this result was not statistically significant after adjusting for BVAB2, M. indolicus, CT, MG, UU, HI and number of recent sex partners, (aOR=0.46, 95% CI=0.16-1.27). There were no significant interactions between the three BVAB in univariate models and no interaction terms were included in adjusted models. Results were similar in a sensitivity analysis restricted to MSW with test results for UU and HI (Supplementary Table S1).
Table 2 –
Association of Detection of BVAB2, M. indolicus, and Sneathia spp. with NGU and Idiopathic NGU Among Cisgender Men who have Sex with Women Attending the PHSKC Sexual Health Clinic (N=310)a
| Characteristics | NGU | Idiopathic NGUc | |||||||
|---|---|---|---|---|---|---|---|---|---|
| OR | 95% CI | p | aORb | 95% CI | p | OR | 95% CI | p | |
| BVAB2 | 0.45 | 0.20-1.03 | 0.06 | 0.59 | 0.14-2.43 | 0.464 | 0.38 | 0.08-1.71 | 0.21 |
| M. indolicus | 0.93 | 0.33-2.58 | 0.89 | 3.40 | 0.68-17.06 | 0.138 | 0.86 | 0.18-4.19 | 0.85 |
| Sneathia spp. | 0.51 | 0.28-0.93 | 0.03 | 0.46 | 0.16-1.27 | 0.134 | 0.48 | 0.18-1.32 | 0.16 |
| C. trachomatis | -- | -- | -- | 24.86 | 7.96-77.68 | <0.001 | -- | -- | -- |
| M. genitalium | -- | -- | -- | 9.98 | 4.03-24.69 | <0.001 | -- | -- | -- |
| U. urealyticum | -- | -- | -- | -- | -- | -- | |||
| Present | -- | -- | -- | 1.64 | 0.70-3.88 | 0.258 | -- | -- | -- |
| Not T ested | -- | -- | -- | 4.78 | 0.47-48.89 | 0.187 | -- | -- | -- |
| H. influenzae | -- | -- | -- | -- | -- | -- | |||
| Present | -- | -- | -- | 27.54 | 3.40-223.11 | 0.002 | -- | -- | -- |
| Not T ested | -- | -- | -- | 0.060 | 0.01-0.62 | 0.018 | -- | -- | -- |
| Sex partners (past 2 months) | -- | -- | -- | 1.46 | 0.79-2.69 | 0.228 | -- | -- | -- |
Seven men were excluded from the model due to missing data on the number of sexual partners in the past 2 months.
Adjusted for all other characteristics shown in the table (the other BVAB, CT, MG, UU, HI, and >1 sex partners in the past two months).
Restricting to men negative for CT, MG, UU, and TV. Only bivariate odds ratios are shown as the number of men with each bacterium was small and resulted in unstable multivariable models
Abbreviations: CI = confidence interval, NGU = nongonococcal urethritis, OR = odds ratio; aOR=adjusted OR
BVAB and Idiopathic NGU
In analyses restricted to men without CT, MG, UU, and TV, there was no significant association of BVAB2 with idiopathic NGU (OR=0.38, 95% CI=0.08-1.71, Table 2), nor was there a significant association with M. indolicus (OR=0.86, 95% CI=0.18-4.19) or Sneathia spp. (OR=0.48, 95% CI=0.18-1.32). Given small numbers, it was not possible to perform multivariable analyses for idiopathic NGU.
Quantity of bacteria and NGU
Adjusting for other organisms and multiple sex partners, quantity of BVAB2 was not associated with NGU (mean log copies/mL 2.61 vs. 2.67, aOR= 0.81, 95% CI=0.44-1.51; Table 3). There was also no association with quantity of M. indolicus, (mean log copies/mL 3.03 vs. 2.71, aOR=1.70, 95% CI=0.94-3.11) or quantity of Sneathia spp. (mean log copies/mL 3.79 vs. 3.82, aOR=0.81, 95% CI=0.63-1.05) and NGU, adjusting for the same confounders. Results were similar when analyses were restricted to MSW with complete test results (Supplementary Table S2).
Table 3 –
Association of Quantity of BVAB2, M. indolicus, and Sneathia spp. (in Log Copies / mL) with NGU Among Cisgender Men who have Sex with Women Attending the PHSKC SHC Clinic (N=310)a
| Characteristics | Unadjusted Analyses | Adjusted Analyses | ||||
|---|---|---|---|---|---|---|
| OR | 95% CI | p-value | aORb | 95% CI | p-value | |
| BVAB2 | 0.80 | 0.61-1.05 | 0.113 | 0.81 | 0.44-1.51 | 0.515 |
| M. indolicus | 1.04 | 0.77-1.41 | 0.791 | 1.70 | 0.94-3.11 | 0.081 |
| Sneathia spp. | 0.85 | 0.73-0.99 | 0.039 | 0.81 | 0.63-1.05 | 0.119 |
| C. trachomatis | -- | -- | -- | 25.14 | 8.05-78.54 | <0.001 |
| M. genitalium | -- | -- | -- | 9.71 | 3.90-24.13 | <0.001 |
| U. urealyticum | -- | -- | -- | |||
| Present | -- | -- | -- | 1.79 | 0.75-4.25 | 0.190 |
| Not tested | -- | -- | -- | 4.81 | 0.47-48.74 | 0.184 |
| H. influenzae | -- | -- | -- | |||
| Present | -- | -- | -- | 29.16 | 3.60-236.34 | 0.002 |
| Not tested | -- | -- | -- | 0.06 | 0.01-0.63 | 0.019 |
| Sex partners (past 2 months) | -- | -- | -- | 1.42 | 0.77-2.63 | 0.260 |
Seven men were excluded from the model due to missing data on number of sexual partners in the past 2 months.
Adjusted for all other characteristics shown in the table (the other BVAB, CT, MG, UU, HI, >1 sex partners in the past two months).
Abbreviations: CI = confidence interval, NGU = nongonococcal urethritis, OR = odds ratio; aOR=adjusted OR
Influence of NGU definition
Under the Objective Signs-Based definition, there was no significant association between NGU and BVAB2 (38.9% vs. 52.3%, p=0.16) or M. indolicus (58.8% vs. 50.3%, p=0.62) and the association between Sneathia spp. and NGU was not statistically significant (40.3% vs. 53.6%, p=0.06). Results were similar among the subset of men without CT, MG, UU, and TV (data not shown). In contrast, under the Symptoms-Based definition, both BVAB2 and Sneathia spp. were significantly less common among men with than without NGU (19.4% vs. 37.7%, p=0.04 and 23.8% vs. 38.8%, p=0.03, respectively). The lack of association of M. indolicus with NGU persisted (35.3% vs. 35.7%, p=1.0). There were no significant associations when analyses were restricted to men without CT, MG, UU, and TV (data not shown).
Natural History
Of 317 included MSW, 109 (34.4%) had qPCR results for BVAB2, M. indolicus and Sneathia spp. for ≥2 follow-up visits. Thirty-four (31.2%) had one or more of these three bacteria detected at one or more follow-up visits and are included on the heatmap (Supplementary Figure S2).
Men frequently had the same bacterium detected over multiple visits and the three BVAB were often detected together. Twenty-two (64.7%) men had >1 BVAB present at one or more visits. All 21 men with BVAB2 at any visit also had Sneathia spp. concurrently detected at least once and only four men (19%) had a time-point where BVAB2 was present alone. M. indolicus was infrequently detected (8 of 34 men) and almost exclusively present together with Sneathia spp. and BVAB2. Sneathia spp. were detected alone more often than the other BVAB (41 of 88 [46.6%] visits at which it was present). We observed no consistent patterns with NGU, PMNs or antibiotic use and detection of any BVAB over time.
Median duration of persistence was 30.5 days (IQR=28-87) for BVAB2, 87 days (IQR=60-126) for M. indolicus, and 70 days (IQR=30-135) for Sneathia spp. and not significantly different between BVAB (p≥0.20 for all). When interim negative tests were reclassified as positive, median persistence was slightly longer, but not significantly different by BVAB: 64 days (IQR =28-138) for BVAB2, 116 days (IQR =87-189) for M. indolicus, and 87 days (IQR =33-178) for Sneathia spp. (all p>0.20).
DISCUSSION
In these cisgender MSW attending a sexual health clinic in Seattle, detection of Sneathia spp. was fairly common (21.1%). BVAB2 was detected in 11.4% and M. indolicus in 5.4%. Aside from detection of UU in a subset of men and a higher number of sex partners in the past two months, few characteristics were associated with these BVAB. While Sneathia spp. were associated with lower odds of prevalent NGU in unadjusted analyses, there was no association between prevalent NGU and either BVAB2, M. indolicus and Sneathia spp. in adjusted analyses. All three BVAB commonly co-occurred and persisted over six-months.
The overall lack of association with sociobehavioral characteristics is consistent with other studies of BVAB in men2,7. Despite the higher prevalence of each of the BVAB in those with ≥3 sex partners in the past two months, we observed no association between having a new partner, suggesting that the BVAB were not recently transmitted. As BV in cisgender women is often persistent and recurrent13,19,20, the association with ≥3 recent sex partners may reflect increased opportunity for exposure to these bacteria from multiple sources.
BV is characterized by significantly increased bacterial diversity, and BVAB frequently co-occur and persist in women21-24. The co-occurrence and persistence of these BVAB that we observed in men parallels observations in women. However, despite increased risk of NGU in MSW whose female partners have BV, male-partner treatment trials have not successfully decreased BV recurrences in women25, although this may be due to methodologic limitations of the trials themselves26-30,S31,S32. Furthermore, not using condoms12 and having a recent new sex partnerS33 have been associated with BV in women, but neither was associated with these BVAB in men. More in-depth study of the transmission dynamics of BVAB are needed to fully understand the role of male-female sexual partnerships.
The association of UU with each of the BVAB is consistent with previously reported associations between UU and BV in cisgender womenS34. Notably, both UU and U. parvum (UP) were associated with 7 of 8 BVAB (including Sneathia spp. and BVAB2, but not M. indolicus) in a case control study of NGU among Swedish men. BVAB organism load was higher in men with higher Ureaplasma organism load (UU+UP)8. Our results are consistent with the hypothesis that pH is elevated in the presence of Ureaplasmas, which facilitates the growth of BVAB.
Few clinical characteristics were associated with these BVAB and we observed no increased odds of prevalent NGU with any of the bacteria. While this is in sharp contrast to our prior study in Seattle men2 it is consistent with the two other studies on this topic7,8. Notably, Sneathia spp. were somewhat less common among Swedish men with idiopathic urethritis than among controls7. This lack of increased risk did not depend on how NGU was defined. Both Swedish studies used slightly different definitions (≥5 PMNs/hpf plus urethritis symptoms8 and ≥10 PMNL/hpf plus urethritis symptoms lasting <30 days7), yet reported consistent results. Our comparison of results under two alternative definitions also demonstrated minimal effect.
This study had several strengths. We used previously validated, highly sensitive and specific assays. All participants were assessed for NGU by the same two clinicians, enhancing the consistency of diagnosis. We assessed multiple definitions of NGU, demonstrating consistency across outcome measures. The longitudinal data on a subset of men allowed us to observe the natural history of these BVAB.
Our study also had several limitations. First, we included men with either clinical signs or symptoms in the absence of elevated PMNs in the comparison group without NGU, as well as men with elevated PMNs alone, potentially attenuating any relationship of these BVAB with NGU. However, our results were consistent with another study that restricted the comparison group to men with neither elevated PMNs nor urethral symptoms8, suggesting that any effect of this is small. Second, the qPCR assay that we used to detect Sneathia spp. did not differentiate between subspecies. It is possible that one species is associated with NGU and other species obscured this. Third, we only had test results for UU, HI, TV, HSV and adenovirus in approximately 55% of our study population, limiting our ability to adjust for these organisms. However, results were similar when the analysis was restricted to the subset of MSW with complete test results and, with the exception of NGU status (less common among MSW with missing data), MSW with and without complete test results were similar. This suggests that the impact of any bias due to missing data is likely to be small. Fourth, while we focused on three BVAB that have been previously associated with NGU, there are many other BVAB. Further studies would be needed to elucidate any potential relationships with these other BVAB and NGU. Finally, our results are specific to cisgender MSW attending a sexual health clinic and may not be generalizable to other populations.
In summary, we were not able to replicate the results of our prior study2. Although these BVAB were frequently persistent and often co-occurred in MSW, they were not associated with increased odds of prevalent NGU. These results do not lend support for testing for these BVAB or for providing targeted treatment for them in the clinical management of urethritis. Regardless of whether these BVAB lead to adverse health outcomes in MSW, they are important risk factors for female sexual health. The consistent association of multiple partners with each of these BVAB should motivate research investigating how the number and variety of female partners influences BVAB acquisition in MSW and subsequent transmission to women. Given the potential for sexual transmission of BVAB12,13, gaining a better understanding of these bacteria in MSW may inform our understanding of acquisition and clearance in women.
Supplementary Material
KEY MESSAGES.
What is already known on this topic:
Nongonococcal urethritis (NGU) is the most common male genital tract syndrome, 40-50% of cases are of unknown etiology, and the role of BV-associated bacteria (BVAB) is controversial.
What this study adds:
There was no evidence that BVAB2, M. indolicus, or Sneathia spp. were associated with increased risk of prevalent NGU in men who have sex with women (MSW) attending a sexual health clinic.
How this study might affect research, practice or policy:
Neither testing nor targeted treatment for BVAB2, M. indolicus or Sneathia spp. should be recommended for MSW with urethritis. Future research should investigate the role of male colonization with these bacteria in the incidence and recurrence of BV in female partners.
Acknowledgements:
The authors thank the study participants and the Public Health – Seattle & King County STD Clinic staff. They also gratefully acknowledge Sean Proll for database support and Dr. Olusegun O. Soge for Mycoplasma genitalium testing, and Hologic, Inc. for donating Aptima test kits and reagents. Study data were collected and managed using REDCap tools hosted at the University of Washington (UW) Institute of Translational Health Sciences. This work was presented in part at the STI and HIV 2021 World Conference, Amsterdam, The Netherlands, 14–17 July 2021. Part of this manuscript was previously published as a thesis:
Zinsli KA, Manhart LE , Balkus JE, et al. Non-Gonococcal Urethritis and Its Relationship to Three Novel Bacterial Vaginosis (BV) Associated Bacteria [Preprint]. 2020 [cited 2022 Aug 23] https://www.proquest.com/dissertations-theses/non-qonococcal-urethritis-relationship-three/docview/2444855799/se-2.S35
Financial Support:
This work was supported by the National Institute of Allergy and Infectious Diseases (R01AI110666 and U19AI113173). Study data were collected and managed using REDCap electronic data capture tools hosted at the Institute of Translational Health Sciences through a grant from the National Center for Advancing Translational Sciences (UL1TR002319).
Footnotes
Conflict of Interest Statement: L.E.M. has received research support and honoraria from Hologic, Inc, and Nabriva Therapeutics.
REFERENCES
- 1.Martin DH. Urethritis in Males. In: Holmes KK, et al. , eds. Sexually Transmitted Diseases. New York, NY: McGraw Hill; 2008: 1107–1126. [Google Scholar]
- 2.Manhart LE, Khosropour CM, Liu C, et al. Bacterial vaginosis-associated bacteria in men: association of Leptotrichia/Sneathia spp. with nongonococcal urethritis. Sex Transm Dis 2013;40:944–949. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Srinivasan S, Chambers LC, Tapia KA, et al. Urethral Microbiota in Men: Association of Haemophilus influenzae and Mycoplasma penetrans With Nongonococcal Urethritis. Clin Infect Dis Off Publ Infect Dis Soc Am 2021;73:e1684–1693. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Ito S, Hatazaki K, Shimuta K, et al. Haemophilus influenzae Isolated from Men With Acute Urethritis: Its Pathogenic Roles, Responses to Antimicrobial Chemotherapies, and Antimicrobial Susceptibilities. Sex Transm Dis 2017;44:205–210. [DOI] [PubMed] [Google Scholar]
- 5.Wetmore CM, Manhart LE, Lowens MS, et al. Ureaplasma urealyticum Is Associated with Nongonococcal Urethritis Among Men With Fewer Lifetime Sexual Partners: A Case-Control Study. J Infect Dis 2011;204:1274–1282. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Bradshaw CS, Tabrizi SN, Read TRH, et al. Etiologies of nongonococcal urethritis: bacteria, viruses, and the association with orogenital exposure. J Infect Dis 2006;193:336–345. [DOI] [PubMed] [Google Scholar]
- 7.Frølund M, Wikström A, Lidbrink P, et al. The bacterial microbiota in first-void urine from men with and without idiopathic urethritis. PloS One 2018;13:e0201380. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Frølund M, Falk L, Ahrens P, et al. Detection of ureaplasmas and bacterial vaginosis associated bacteria and their association with non-gonococcal urethritis in men. PloS One 2019;14:e0214425. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Koumans EH, Sternberg M, Bruce C, et al. The prevalence of bacterial vaginosis in the United States, 2001-2004; associations with symptoms, sexual behaviors, and reproductive health. Sex Transm Dis 2007;34:864–869. [DOI] [PubMed] [Google Scholar]
- 10.Fredricks DN, Fiedler TL, Marrazzo JM. Molecular identification of bacteria associated with bacterial vaginosis. N Engl J Med 2005;353:1899–1911. [DOI] [PubMed] [Google Scholar]
- 11.Gorgos LM, Sycuro LK, Srinivasan S, et al. Relationship of Specific Bacteria in the Cervical and Vaginal Microbiotas with Cervicitis. Sex Transm Dis 2015;42:475–481. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Sanchez S, Garcia PJ, Thomas KK, et al. Intravaginal metronidazole gel versus metronidazole plus nystatin ovules for bacterial vaginosis: a randomized controlled trial. Am J Obstet Gynecol 2004;191:1898–1906. [DOI] [PubMed] [Google Scholar]
- 13.Marrazzo JM, Thomas KK, Ringwood K. A behavioural intervention to reduce persistence of bacterial vaginosis among women who report sex with women: results of a randomised trial. Sex Transm Infect 2011;87:399–405. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Chambers LC, Morgan JL, Lowens MS, et al. Cross-sectional study of urethral exposures at last sexual episode associated with non-gonococcal urethritis among STD clinic patients. Sex Transm Infect 2019;95:212. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Corey L, Huang M-L, Selke S, et al. Differentiation of herpes simplex virus types 1 and 2 in clinical samples by a real-time taqman PCR assay. J Med Virol 2005;76:350–355. [DOI] [PubMed] [Google Scholar]
- 16.Kuypers J, Wright N, Ferrenberg J, et al. Comparison of Real-Time PCR Assays with Fluorescent-Antibody Assays for Diagnosis of Respiratory Virus Infections in Children. J Clin Microbiol 2006;44:2382–2388. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Fredricks DN, Fiedler TL, Thomas KK, et al. Changes in vaginal bacterial concentrations with intravaginal metronidazole therapy for bacterial vaginosis as assessed by quantitative PCR. J Clin Microbiol 2009;47:721–726. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Rowlinson E, Hughes JP, Chambers LC, et al. Incidence of Nongonococcal Urethritis in Men Who Have Sex With Women and Associated Risk Factors. Sex Transm Dis 2021;48:341–346. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Spiegel CA. Bacterial vaginosis. Clin Microbiol Rev 1991;4:485–502. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Workowski KA, Bolan GA. Sexually Transmitted Diseases Treatment Guidelines, 2015. MMWR Recomm Rep Morb Mortal Wkiy Rep Recomm Rep 2015;64:1–137. [PMC free article] [PubMed] [Google Scholar]
- 21.Srinivasan S, Hoffman NG, Morgan MT, et al. Bacterial Communities in Women with Bacterial Vaginosis: High Resolution Phylogenetic Analyses Reveal Relationships of Microbiota to Clinical Criteria. PLoS One 2012;7:e37818. doi: 10.1371/journal.pone.0037818. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Barrientos-Duran A, Fuentes-López A, de Salazar A, et al. Reviewing the Composition of Vaginal Microbiota: Inclusion of Nutrition and Probiotic Factors in the Maintenance of Eubiosis. Nutrients 2020;12:419. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Oakley BB, Fiedler TL, Marrazzo JM, et al. Diversity of Human Vaginal Bacterial Communities and Associations with Clinically Defined Bacterial Vaginosis. Appl Environ Microbiol 2008;74:4898–909. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Hillier SL MJ, Holmes KK. Bacterial vaginosis. In: Holmes KK, Sparling PF, Mardh PA, eds. Sexually Transmitted Diseases. New York, NY: McGraw Hill; 2008:737–768. [Google Scholar]
- 25.Keane FE, Thomas BJ, Whitaker L, et al. An association between non-gonococcal urethritis and bacterial vaginosis and the implications for patients and their sexual partners. Genitourin Med 1997;73:373–377. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Vejtorp M, Bollerup AC, Vejtorp L, et al. Bacterial vaginosis: a double-blind randomized trial of the effect of treatment of the sexual partner. BrJ Obstet Gynaecol 1988;95:920–926. [DOI] [PubMed] [Google Scholar]
- 27.Swedberg J, Steiner JF, Deiss F, et al. Comparison of single-dose vs one-week course of metronidazole for symptomatic bacterial vaginosis. JAMA 1985;254:1046–1049. [PubMed] [Google Scholar]
- 28.Mengel MB, Berg AO, Weaver CH, et al. The effectiveness of single-dose metronidazole therapy for patients and their partners with bacterial vaginosis. J Fam Pract 1989;28:163–171. [PubMed] [Google Scholar]
- 29.Colli E, Landoni M, Parazzini F. Treatment of male partners and recurrence of bacterial vaginosis: a randomised trial. Genitourin Med 1997;73:267–270. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Vodstrcil LA, Muzny CA, Plummer EL, et al. Bacterial vaginosis: drivers of recurrence and challenges and opportunities in partner treatment. BMC Med 2021;19:194. doi: 10.1186/s12916-021-02077-3 [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.