Objective: To compare pregnancy outcomes using self-reported and objective levels of intracellular tenofovir diphosphate (TFV-DP) in pregnant women using preexposure prophylaxis (PrEP).

Design: We enrolled pregnant women >15 years without HIV at first antenatal care visit in an observational cohort study to compare pregnancy outcomes by PrEP use.

Methods: Exposure defined as: any PrEP use [tenofovir disoproxil and emtricitabine (TDF/FTC]) prescription + reported taking PrEP], or objectively-measured TFV-DP in dried blood spots in PrEP-using pregnant women. The primary outcome was a composite of pregnancy loss, preterm birth (<37weeks), low birthweight (<2500 g), small for gestational age ([SGA] ≤ tenth percentile), or neonatal death. Multivariable logistic regression models evaluated individual and composite adverse outcomes by self-reported or objectively measured PrEP use adjusting for age, gestational age, gravidity and socio-economic status.

Results: Between August 19 and February 23, we followed 1195 pregnant women and ascertained 1145 pregnancy outcomes (96%); 72% ( n = 826) reported taking PrEP while pregnant, 16% did not take PrEP ( n = 178), 12% were unconfirmed ( n = 141). Overall, 94.5% ( n = 1082) had singleton live births with a median birthweight of 3.2 kg [interquartile range (IQR) = 2.9-3.5], with no difference in pregnancy loss between self-reported PrEP exposed vs. unexposed [4.0 vs. 5.6%; adjusted odds ratio (aOR) = 0.65, 95% confidence interval (CI) = 0.32-1.47]. Composite adverse outcomes did not differ by reported PrEP use (20% for both groups; aOR = 1.07, 95% CI = 0.71-1.63). Comparing objective PrEP use (any TFV-DP vs. no TFV-DP or not on PrEP), adverse outcomes did not differ (aOR = 0.64, 95% CI = 0.39-1.04), nor did other outcomes including preterm birth nor SGA.

Conclusions: Pregnancy outcomes did not differ by PrEP exposure (self-reported or objective), suggesting real-world efficacy that TDF/FTC as PrEP is safe in pregnancy.