OBJECTIVE: The purpose of this study was to investigate first trimester anticoagulant exposure and risks of adverse pregnancy-related and fetal outcomes.
METHODS: Using Danish nationwide registries, we identified all pregnant women with preconception venous thromboembolism, 2000-2017, and linked data on exposure to low-molecular-weight heparin (LMWH), vitamin K antagonist (VKA), or non-VKA oral anticoagulant (NOAC) during pregnancy. We assessed pregnancy-related and fetal outcomes associated with first trimester anticoagulant exposure.
RESULTS: Among 4490 pregnancies in women with preconception venous thromboembolism (mean age 31 years, 40% nulliparous) during the first trimester, 63.1% were unexposed, 25.9% were exposed to LMWH, 10.4% VKA, and 0.6% NOAC. Adverse outcomes were lowest in unexposed and LMWH exposed. Compared with unexposed, VKA was associated with higher risks of preterm (adjusted odds ratio [OR] 2.26; 95% confidence interval [CI], 1.70-2.99) and very preterm birth (adjusted OR 3.78; 95% CI, 1.91-7.49), shorter mean gestational age was associated with VKA (-7.5 days; 95% CI, -9.1 to -5.9 days) or NOAC (-2.3 days; 95% CI, -8.4-3.8), and lower mean birthweight with VKA (-55 g; 95% CI, -103.1 to -8.5) or NOAC (-190 g; 95% CI, -364.1 to -16.4). Adjusted ORs for small-for-gestational-age infants were 1.07 (95% CI, 0.77-1.50) with VKA, and 3.29 (95% CI, 1.26-7.95) with NOAC. Mean 5-minute Apgar score (9.8) and congenital defect prevalence (8.4%-10%) varied little across exposure groups.
CONCLUSIONS: Fetal risk was lowest in unexposed and LMWH-exposed pregnancies, supporting the recommendation of LMWH during pregnancy. NOAC safety during pregnancy is unclear due to the rarity of NOAC exposure.
This is very consistent with what we already know about the safety of LMWH in pregnancy and the risks for VKA. Unfortunately, due to the small exposed population, it doesn't add much to the area of greatest need/uncertainty: risks with DOAC exposure.
The results are interesting, however, it does not demonstrate anything new. The questions that remain unanswered are: 1. Should we treat all women who have had a previous VTE with anticoagulants?; 2. What is the most effective way to recognize pregnancy as early as possible in a woman on anticoagulation therapy? Would systematic screening make sense?; 3. Are all DOACs dangerous in pregnancy?; 4. Might the new activated factor XI inhibitors have a rationale in this population? (especially antisense oligonucleotides). A study that can answer one of these questions would be able to change clinical practice significantly.