Combined oral and topical antimicrobial therapy for male partners of women with bacterial vaginosis: Acceptability, tolerability and impact on the genital microbiota of couples - A pilot study

Abstract

Objectives: Recurrence following recommended treatment for bacterial vaginosis is unacceptably high. While the pathogenesis of recurrence is not well understood, recent evidence indicates re-infection from sexual partners is likely to play a role. The aim of this study was to assess the acceptability and tolerability of topical and oral antimicrobial therapy in male partners of women with bacterial vaginosis (BV), and to investigate the impact of dual-partner treatment on the vaginal and penile microbiota.

Methods: Women with symptomatic BV (Nugent Score of 4-10 and ≥3 Amsel criteria) and their regular male sexual partner were recruited from Melbourne Sexual Health Centre, Melbourne, Australia. Women received oral metronidazole 400mg twice daily (or intra-vaginal 2% clindamycin cream, if contraindicated) for 7-days. Male partners received oral metronidazole 400mg twice daily and 2% clindamycin cream topically to the penile skin twice daily for 7-days. Couples provided self-collected genital specimens and completed questionnaires at enrolment and then weekly for 4-weeks. Genital microbiota composition was determined by 16S rRNA gene sequencing. Changes in genital microbiota composition were assessed by Bray-Curtis index. Bacterial diversity was measured by the Shannon Diversity Index.

Results: Twenty-two couples were recruited. Sixteen couples (76%) completed all study procedures. Adherence was high; most participants took >90% of prescribed medication. Medication, and particularly topical clindamycin in males, was well tolerated. Dual-partner treatment had an immediate and sustained effect on the composition of vaginal microbiota (median Bray-Curtis score day 0 versus day 8 = 0.03 [IQR 0-0.15], day 0 vs day 28 = 0.03 [0.02-0.11]). We observed a reduction in bacterial diversity of the vaginal microbiota and a decrease in the prevalence and abundance of BV-associated bacteria following treatment. Treatment had an immediate effect on the composition of the cutaneous penile microbiota (median Bray-Curtis score day 0 vs day 8 = 0.09 [0.04-0.17]), however this was not as pronounced at day 28 (median Bray-Curtis score day 0 vs day 28 = 0.38 [0.11-0.59]). A decrease in the prevalence and abundance of BV-associated bacteria in the cutaneous penile microbiota was observed immediately following treatment at day 8.

Conclusion: Combined oral and topical treatment of male partners of women with BV is acceptable and well tolerated. The combined acceptability and microbiological data presented in this paper supports the need for larger studies with longer follow up to characterize the sustained effect of dual partner treatment on the genital microbiota of couples and assess the impact on BV recurrence.

Fig 1. Participant flowchart.

Participant flowchart detailing number of women screened for eligibility, resulting number of couples recruited to study and their progression through the study period. LTFU, lost to follow up.

Fig 2. Specimen flowchart.

Specimen flowchart detailing number of vaginal and penile skin specimens available for microbiota analysis at baseline, day 8 and day 28. Seventeen women provided vaginal specimens for day 0 and 8 paired comparisons, and 16 provided vaginal specimens for day 0 and 28 paired comparisons. Sixteen males provided cutaneous penile specimens for day 0 and 8 paired comparisons and 15 males provided cutaneous penile specimens for day 0 and 28 paired comparisons. The number of couples providing specimens at each time-point is also shown. ^abaseline specimen was not available for one female; ^btwo d8 penile skin specimens failed to meet the sequence depth threshold and were substituted with day 14 specimens.

Fig 3. Heatmap of bacterial abundance from vaginal specimens collected at baseline, day 8 and day 28.

Each vertical line represents the bacterial composition of one vaginal specimen. Only the 30 most abundant taxa found in vaginal specimens are included in the heatmap. Study day is displayed above the heatmap in red (day 0), blue (day 8) and yellow (day 28). Specimens collected from females who experienced BV recurrence during the study are indicated by * and # below the dendrogram.

Fig 4. Heatmap of bacterial abundance from penile skin specimens collected at baseline, day 8 and day 28.

Each vertical line represents the bacterial composition of one penile specimen. Only the 30 most abundant taxa found in penile specimens are included in the heatmap. Study day is displayed above the heatmap in red (day 0), blue (day 8) and yellow (day 28); circumcision status is displayed in black (uncircumcised) and grey (circumcised). Specimens collected from male partners of women who experienced BV recurrence during the study are indicated by * and # below the dendrogram.

Fig 5. Immediate and sustained effect of dual partner treatment on the composition and diversity of the genital microbiota of females and males.

Panel A. Effect of treatment on microbiota composition. Bray-Curtis scores were calculated using the between paired specimens from each participant to investigate the change in microbiota composition from baseline to day 8 (D0 vs D8) and baseline to day 28 (D0 vs D28). A lower Bray-Curtis score indicates greater change in microbiota composition. Panel B. Effect of treatment on microbiota diversity. Alpha diversity is expressed as effective number of taxa, which is defined as the exponent of the Shannon Diversity Index. Alpha diversity values are presented for specimens collected at baseline (D0), day 8 (D8) and day 28 (D28). Changes in alpha diversity between baseline and day 8 and baseline and day 28 were assessed by the Wilcoxon signed-rank test. Box and whisker plots show median, Interquartile range (IQR), and the most extreme values within 1.5 IQR of the nearest quartile (dots indicate outliers).