The Effect of Isoniazid for Tuberculosis Prevention on Pregnancy Risk Among Women Living With HIV on Antiretroviral Treatment and Progestin-Based Hormonal Contraception.

Purpose: Concomitant use of antiretroviral therapy (ART), hormonal contraception, and isonicotinic acid hydrazide (isoniazid) for tuberculosis prevention is common among women of reproductive age who are living with HIV in sub-Saharan Africa. We estimated the effect of isoniazid on 6-month pregnancy risk among Malawian women living with HIV in the Family Planning and Antiretroviral Therapy (FP-ART) prospective cohort study, overall and among subgroups defined by ART regimen type and hormonal contraceptive method.

Methods: The analytic sample included visits contributed by participants who were currently using either efavirenz- or dolutegravir-based ART and either depot medroxyprogesterone acetate (DMPA) or levonorgestrel (LNG) implant contraception at the time of the visit. The exposure was self-reported, current isoniazid use (yes/no). The binary outcome measure, 6-month pregnancy, was defined as an estimated conception date 1-183 days after the study visit date. We used a marginal structural linear risk regression model with inverse probability of treatment weights, multiple imputation by chained equations, and bootstrapping to estimate risk differences (RD) and 95% confidence intervals (CI).

Results: The analytic sample included 4709 study visits occurring between September 2017 and June 2021. The weighted 6-month pregnancy risk among isoniazid use visits was 3.0% compared with 2.3% among non-use visits (RD 0.7%, 95% CI: -0.7%, 2.1%), and the results were qualitatively similar for all subgroup estimates.

Conclusions: We did not find a clinically significant effect of isoniazid use on 6-month pregnancy incidence among women concomitantly using ART and either DMPA or LNG implant contraception.