Abstract
PURPOSE: Doxycycline is frequently prescribed during pregnancy, yet evidence on fetal safety is inconsistent and often excludes non-live births. We assessed whether exposure during the first or third trimester is associated with major congenital malformations or late-pregnancy adverse outcomes in a population-based cohort that also included stillbirths and terminations. METHODS: Using data from Clalit Health Services Southern district, we identified 265,686 pregnancies in women aged 15-45 years (from 1998 to 2017). Pharmacy records classified doxycycline dispensation in the first trimester (≤ 13 weeks) or third trimester (≥ 27 weeks). Crude and adjusted negative-binomial models estimated relative risks (RRs) for total and organ-specific major congenital malformations diagnosed up to age 1 year and for perinatal mortality, preterm birth, low/very-low birthweight, and low Apgar scores. Sensitivity analyses explored dose-response relations and propensity-score-matched cohorts. RESULTS: Among 2,696 first-trimester exposures, major malformations occurred in 7.7% versus 7.0% of 262,990 unexposed pregnancies (SMD = 0.03, p = 0.17). No association with major malformations was observed in both crude (Crude Relative Risk (RR) = 1.10; 95% CI 0.96-1.27) and adjusted (Adjusted RR = 1.07; 95% CI 0.93-1.23) analyses, nor by organ-specific sub-groups. Third-trimester exposure (n = 112) was linked to a higher risk of very-low birthweight, while other late-pregnancy outcomes were comparable to unexposed pregnancies. CONCLUSION: First-trimester doxycycline use was not associated with increased major congenital malformation risk, and most late-pregnancy outcomes were unaffected. These findings support the relative safety of doxycycline when clinically indicated during pregnancy.