Effect of contraceptive methods on the vaginal microbiome and host immune factors

Abstract

Objective: The objective of this study was to assess alterations in vaginal microbiota and immune markers over the first 3 months following initiation of copper intrauterine device (copper IUD), levonorgestrel (LNG) implant, and intramuscular depot medroxyprogestone acetate (DMPA-IM).

Study design: We included 162 participants from the Evidence for Contraceptive Options and HIV Outcomes (ECHO) trial, which enrolled healthy, HIV-negative women seeking contraception and randomized them to a copper IUD, LNG implant, or DMPA-IM. Microbiome and immune profiles in vaginal swab samples collected at enrollment, 1 month and 3 months were analyzed. We categorized microbiome profiles as ''optimal'', ''intermediate'', or ''non-optimal'' based on established criteria [1]. We compared microbiome and immune markers across contraceptive groups and evaluated changes to 1 and 3 months.

Results: Copper IUD users had a more diverse vaginal microbiome and generally increased inflammatory cytokines and antimicrobial peptides over the 3-month follow-up, compared to LNG-implant and DMPA-IM users [2]. LNG-implant users had less complex vaginal microbiomes with reduced inflammation, while DMPA-IM showed little change in either microbiome composition or inflammatory markers. Copper IUD users exhibited lower microbiome stability and a higher likelihood of transitioning to less optimal profiles. In contrast, LNG-implant users showed greater stability and a higher probability of transition to optimal microbiome and immune marker profiles.

Conclusions: Contraceptive methods affect the vaginal microbiome differently. Copper IUD use may lead to less favorable profiles and increased levels of some immune markers, indicating potential adverse health effects. Conversely, LNG-implant usage promotes a more favorable microbiome and immune marker balance.

Implications: Our findings suggest that copper IUDs are associated with decreased prevalence of Lactobacillus-dominated microbiomes, higher transition rates towards less optimal microbiome and increased inflammatory profiles, which may lead to negative implications for gynecologic and reproductive health, the LNG-implant may offer positive health benefits with increased prevalence of L. crispatus-dominated microbiomes.

Keywords: Copper IUD; DMPA-IM; Inflammation; LNG-implant; Lactobacillus; Vaginal microbiome.

Fig. 1.

Transitions over time of vagitypes in response to different contraceptive methods (Tshwane and in the eThekwini District, South Africa, 2015–2017). (A) Alpha diversity was quantified by calculating the Shannon index. (B) Evenness of taxa distribution. (C) The number of observed taxa. * Indicates p = 0.05 and ** p = 0.01 after adjusting for multiple comparisons. (D) Longitudinal changes in Optimal, Intermediate and Non-optimal vaginal microbiome. (E) Vaginal microbiome counts across time with respect to contraceptive method.

Fig. 2.

Vaginal microbiome transitions in women using different contraceptive methods (Tshwane and in the eThekwini District, South Africa, 2015–2017). We assessed the probability of transition of the vaginal microbiomes in each contraceptive group longitudinally across the three month period using a simplified Markov model. (A) Alluvial diagram illustrating the transitions among vaginal microbiome profiles among women using different contraceptive methods. The numbers of women in each transitional group are indicated. (B) Transition frequencies among women using different contraceptive methods.

Fig. 3.

Effect of contraceptive method on cytokine and antimicrobial peptide expression levels (Tshwane and in the eThekwini District, South Africa, 2015–2017). A mixed effect model was fit using JMP version 16 to each cytokine and antimicrobial peptide including a random subject effect and covariates for (A) Contraceptive method (copper IUD, DMPA-IM, LNG-implant), visit (baseline, M1 or M3), BMI, active STI, and (B) Vaginal microbiome profile (Optimal, Intermediate, Nonoptimal). BMI = body mass index; DMPA-IM = intramuscular depot medroxyprogesterone acetate; IUD = intrauterine device; LNG = levonorgestrel; STI = sexually transmitted infection.

Fig. 4.

Changes in cytokine, chemokine and antimicrobial peptide levels (Tshwane and in the eThekwini District, South Africa, 2015–2017). Transitions among Optimal (O), Intermediate (I) and Nonoptimal (N) vaginal microbiome profiles at month 1 (A) and month 3 (B).