IAS 2019 — The vaginal microbiome is linked to hormonal contraception-related HIV risk

  • Laura Vargas-Parada, Ph.D.
  • Conference Reports
Access to the full content of this site is available only to registered healthcare professionals. Access to the full content of this site is available only to registered healthcare professionals.


  • The vaginal microbiome is linked to inflammation and rates of HIV incidence in women using different hormonal contraceptives. 
  • Injectable hormonal contraceptive depot medroxyprogesterone acetate (DMPA)-associated inflammation is higher with Lactobacillus-dominant microbiome.

Why this matters 

  • Genital inflammation affects HIV infection risk.
  • Both DMPA and the vaginal microbiome are tied to effects on genital inflammation.
  • DMPA is associated with increased risk of HIV-1 acquisition in women.

Study design 

  • HIV prevention trial in South Africa. 
  • Mass spectrometry-based proteomics analysis:
    • 61 cervicovaginal mucosal specimens from women who acquired HIV during the trial (cases).
    • 624 available women who remained uninfected (controls).

Key results  

  • 97.7% women were using hormonal contraceptives: 
    • DMPA (65.6%).
    • Norethisterone enanthate (NET-EN) (18.0%).
    • Combined oral contraceptives (COC) (14.2%). 
  • 2 major vaginal microbiome profiles detected:
    • Predominant Lactobacillus species (59.4%).
    • Non-Lactobacillus-dominant (40.6%).
  • Microbiome groups were similarly distributed among contraceptive types. 
  • Probability of HIV infection did not differ between DMPA compared with NET-EN/COC use (case-control analysis OR: 1.56; 95% CI, 0.87-2.95; P=.151). 
  • Non-Lactobacillus-dominant women's risk of HIV acquisition not significantly higher in those using DMPA compared with all other hormonal contraceptives (OR: 0.95; 95% CI, 0.44-2.15; P=.895). 
  • With Lactobacillus-dominant microbiome, DMPA use was associated with a >3-fold increase in HIV acquisition risk relative to other hormonal contraceptives (OR: 3.27; 95% CI, 1.24-11.24; P=.0305).
  • Interaction analysis suggested a statistical trend toward the vaginal microbiome having an effect on DMPA-associated HIV risk (P=.0686). 


  • Limited in statistical power, restricted to South African population, and presented in a conference without peer review.

Expert comment 

  • “An important question to answer is, if we stratified by microbiome types, do we see that DMPA modulates inflammation differently?” said Adam Burgener, head of Proteomics at National HIV and Retrovirology Labs at the Public Health Agency of Canada, Associate Professor at the University of Manitoba and speaker at the plenary session at IAS 2019.