BACKGROUND Pregnancy is associated with a hypercoagulable state, therefore the optimal anticoagulants for potential use in pregnant women with prosthetic heart valves are controversial. The aim of this study is to investigate the effect of anticoagulants on pregnancy outcomes and their potential risks in pregnant women with mechanical heart valves. MATERIALS AND METHODS In this prospective cohort study, we followed 44 women with 49 pregnancies who had mechanical heart valves from September 2002 to September 2007. A total of 38 patients took warfarin throughout their pregnancies (group A). In 11 patients, warfarin was changed to heparin during the first trimester and then again to warfarin during 12thto 36thweeks of gestational age (group B). All women took warfarin from 36thweeks of gestational age until delivery. RESULTS In group A, there were 22 live births (57.9%), 15 abortions (39.5%) and 1 maternal death (2.6%). In group B, there were seven live births (63.6%), three spontaneous abortions (27.3%) and one intra-uterine fetal death (9.1%). There was no significant difference in live birth rate between the two groups (p=0.24). Thirty-three pregnancies (86.8%) in group A and five pregnancies (45.4%) in group B had no maternal complications (p=0.004). The difference in pregnancy complications between both groups was significant (p<0.001). CONCLUSION The present study shows that low dose warfarin (5 mg/day or less) may be safe during the first trimester of pregnancy. Maternal adverse events are low when pregnant women with mechanical heart valves remain on a warfarin regimen. The risk of embryopathy does not necessarily increase.