HIV-infected pregnant women have high rates of preterm birth (PTB). Multiple recent reports have suggested, counterintuitively, that women who start antiretroviral therapy (ART) prior to conception have higher rates of PTB than women who initiate ART in pregnancy. We hypothesized that this association, if real, is not mediated through inflammation and immune activation in the vagina and peripheral blood. We analyzed specimens from an ongoing prospective cohort study in Lusaka, Zambia. At the time of analysis, paired (vaginal, plasma) mid-trimester (16-20 weeks) samples from 494 women were available. To maximize information with available budget for this sub-study, we analyzed all specimens from HIV seropositive women and a random sample of HIV negative women, accounting for this with probability weighting in analyses. We further divided HIV-infected women into those on preconceptional ART versus those who started ART during pregnancy (after samples obtained). We measured a panel of 13 maternal plasma and 14 maternal vaginal fluid analytes using a multiplex immunoassay (MilliporeSigma, Darmstadt, Germany). We log-transformed concentrations and compared them across exposure categories with ANOVA. We analyzed 244 plasma specimens (153 HIV-uninfected, 51 HIV-infected on preconceptional ART, and 40 HIV- infected without preconceptional ART exposure) and 255 vaginal swabs (160 HIV-uninfected, 54 HIV-infected on preconceptional ART, and 41 HIV-infected without preconceptional ART exposure). When compared to HIV- uninfected women and HIV-infected women on preconceptional ART, women not on preconceptional ART had significantly higher concentrations of inflammatory plasma (IP10, IL-8, IFN-ɣ, IL-12p40, and sCD14) and vaginal (IL-1β, IL-10, sCD14) biomarkers as well as significantly lower concentrations of anti-inflammatory TGF-β in plasma and vaginal SLPI, an enzyme with anti-HIV properties (p< 0.05; no adjustment for multiple comparisons; Table). HIV-infected pregnant women without preconceptional ART exposure exhibited increased systemic and local inflammation in the second trimester of pregnancy. If recent reports associating preconceptional ART with PTB are correct, our study suggests the mechanism is not early local or systemic inflammation. Additional analyses to understand the interaction between these inflammatory markers, other PTB risk factors, and the vaginal microbiota are ongoing.
Read full abstract