Coumarin derivatives are the anticoagulants most widely used in the United States. These agents are relatively contraindicated during pregnancy, and the use of these drugs in breast-feeding women remains controversial. Much of the confusion regarding the passage of these agents into breast milk might stem from the fact that different agents possess significantly different chemical properties. A review of the chemical structure of different coumarin derivatives, as well as available clinical evidence, suggests that warfarin sodium is not excreted into breast milk, and can be safely given to women requiring therapeutic anticoagulation postpartum. For the rare patient who cannot tolerate warfarin sodium, the use of dicumarol, rather than anisindione, is preferred.
Coumarin derivatives (and the related inandione derivatives) currently are the most widely used agents for chronic anticoagulation in nonpregnant patients. Dicumarol was initially identified in 1939 as the bioactive agent in fermented sweet clover, which caused hemorrhage in cattle.1 In 1948, warfarin, a more potent synthetic congener of dicumarol was developed under the auspices of the Wisconsin Alumni Research Foundation (hence the acronym WARF) as a rodenticide. However, this agent was not used clinically until after 1951, following treatment of a Korean War army inductee who had attempted suicide with the rodenticide warfarin and the ensuing observation of the potential nonlethal use of this agent as an anticoagulant.2
The hazards of coumarin derivatives during pregnancy are well known; such agents are among the most potentially fetotoxic drugs available for general clinical use today. Use during the first trimester results in up to a 10–25% incidence of fetal warfarin syndrome, and exposure of the fetus at later stages of gestation might give rise to varying degrees of fetal cerebral hemorrhage, with subsequent scarring and neurologic abnormalities.3,4 Although the exact incidence of the latter complication is debated, it has been estimated that as many as 5% of infants exposed to these agents after the first trimester will be affected in this manner.4
Despite a paucity of data suggesting harmful effects of the most commonly available coumarin derivatives to breast-feeding infants, many obstetricians remain reluctant to prescribe these drugs to breast-feeding mothers. Indeed, the most recent edition of a widely read obstetric text describes the use of warfarin by breast-feeding mothers as controversial.5 These concerns might represent, in part, extrapolations from the known harmful effects of such agents on the fetus during pregnancy, as well as confusion regarding the widely disparate effects of various coumarin derivatives, some of which are no longer available for clinical use in the United States.
The purpose of this review is to summarize the known effects of maternally administered coumarin and inandione derivatives on breast-feeding infants. Variations in biologic effects might be related to readily identifiable differences in chemical structure.