IntroductionMother-to-child-transmission (MTCT) of human immunodeficiency virus (HIV) can occur in pregnancy/in utero (IU), during childbirth/intrapartum (IP), or postpartum (PP) through breastfeeding from an infected mother to her infant. Burden of PP-MTCT and associated risk factors remain poorly described, especially in adolescent girls and young women (AGYW) aged 15–24 years. Furthermore, despite concerns on high postnatal seroconversions, there is paucity of data on the burden of subsequent MTCT rates.MethodsPregnant women ≥20 weeks of gestation were enrolled into the University of Zimbabwe Birth Cohort from four primary health centers in Harare, Zimbabwe. Mother–infant dyads were followed up from delivery, week(s) 1, 6, 10, 14, 24, 36, 48, 72, and 96 after birth. Women who were uninfected at baseline were re-tested for HIV on subsequent visits. Plasma HIV RNA was quantified using reverse transcriptase polymerase chain reaction. Exposed babies were tested for HIV using qualitative/quantitative proviral DNA PCR on dried blood spots. Maternal–infant factors were tested in univariable/multivariable regression analyses for HIV-MTCT predictors.ResultsA total of 600 HIV-uninfected and 608 HIV-infected pregnant women on Tenofovir/Lamivudine/Efavirenz regimen were enrolled from 2016 to 2019. Postnatal HIV incidence was 0.42 cases/100 women-years [95% confidence interval (CI): 0.12–1.1]. Postnatal seroconverters were less likely to have children/pregnancies sharing same father and unaware of their spouses/intimate partner’s HIV status: p = 0.008 and p = 0.02, respectively, compared with non-seroconverters.Overall HIV-MTCT rate was (15/549): 2.7% (CI: 1.3–4.1%); (7/93) 7.5% observed in AGYW against 1.7%; in women aged >24, p = 0.008. PP-MTCT was the predominant 9/15 (60%) route, followed by IP-MTCT 4/15 (26.6%), whereas IU and postnatal MTCT rates each contributed 6.7% of all infant infections. Postnatal MTCT incidence was 12.8 (CI: 0.3–71.4) infant HIV infections/100 child-years of breastfeeding; a rate 14 times higher than PP-MTCT rate in babies born to women HIV-infected pre/post-conception whose babies were HIV DNA PCR–negative at six weeks.Antenatal HIV RNA >1,000 copies/ml was independently associated with MTCT; odds ratio [CI: 9.3 (2.6–43.1)]. Infected infants’ pre–HIV treatment HIV RNA levels correlated positively with maternal viral load; Spearman’s rank correlation. r = 0.6; p = 0.03.DiscussionMothers were 9.3 times more likely to transmit if HIV RNA was >1,000 copies/ml, disproportionately occurring in vulnerable AGYW. Breastfeeding-associated PP-MTCT remains high; therefore, it is imperative that HIV-infected women commence antiretroviral therapy early in pregnancy to suppress HIV RNA until weaning to decrease the risk of MTCT and possibly reduce the severity of disease in infected infants. HIV-uninfected lactating mothers should be continuously counseled on the risks of postnatal seroconversion.www.clinicaltrials.gov, trial registration number: NCT04087239.