Bacterial Vaginosis Prior to and During First Pregnancy in Kenyan Adolescent Girls and Young Women at Risk Of HIV

Thesis (Master's)--University of Washington, 2021Background: There is limited understanding of longitudinal changes brought about by bacterial vaginosis (BV) in the female genital tract of adolescent girls and young women (AGYW) at pregnancy. AGYW are at a high risk of HIV, STI and pregnancy soon after first sex. Hormonal shifts appear to influence vaginal dysbiosis; a pregnancy at a young age might disproportionately affect the vaginal environment to increase risk of HIV. The aim of this study is to determine the change in BV status of AGYW at preconception and during pregnancy. Using precisely collected longitudinal cohort data, this study examined how vaginal dysbiosis in recently sexually active young women is changing at the time of a first early pregnancy.Methods: We conducted a secondary data analysis of the Girls’ Health Study (GHS) cohort data from 2014 to 2020. Nugent scoring of vaginal Gram stains was used to diagnose BV. A score of 7 and above was considered positive for BV. Logistic regression models were used to analyze longitudinal trends in BV over time, and to examine whether there is increased risk of BV at visits during pregnancy compared to visits before pregnancy. Relevant covariates were adjusted for, including socioeconomic status, marital status, sexual history and reproductive history. Time-to-event analysis was used to describe timing of pregnancy. Additionally, Cox regression was used to assess associations between correlates and pregnancy. Results: We enrolled 400 AGYW, aged 16-20 years, median age 18.6 years (17.6-19.4) into the study, and followed them for a median of 51 months (IQR: 27-57). At the end of follow-up, 306 (76%) had reported first penile-vaginal sex; median age of first sex was 18.9 years (Interquartile range (IQR): 18.3 - 19.9). Forty-two percent (127/306) of sexually active AGYW had a positive pregnancy test at least once during follow up. Contraception use among participants who ever reported pregnancy was low at 26% and only 9/127 (7.1%) reported condom use. The percentage of participants with BV before pregnancy was 38% (45/119) and during pregnancy 23% (24/105). The adjusted relative risk (aRR) of BV during pregnancy among AGYW who had experienced BV pre-pregnancy was 0.66 (95% CI : 0.48, 0.92; p value= 0.015). Factors that were associated with BV during pregnancy included history of CT infection (RR:4.13; 95%CI: 1.73-9.90; p value=0.001). Median number of sex acts increased during pregnancy: 8 sex acts (IQR 2-23) (timepoint defined as LMP + 45 days). Number of sex acts was not associated with increased BV in pregnancy. Conclusions: BV was noted to be more prevalent pre-pregnancy compared to during pregnancy, and pregnancy was associated with a near 40% reduction in BV diagnosis. . Further understanding of the vaginal environment and vaginal bacteria and before and during pregnancy is needed to assess the specific factors that might contribute to reduced BV risk among pregnant AGYW