Estrogens, both endogenous and exogenous, lower the fasting levels of the independent risk factor for cardiovascular disease homocysteine. The mechanism behind this observation remains unclear. In a randomized, placebo-controlled, double-blind study, 25 postmenopausal women with a screening homocysteine concentration above 10 micromol/liter were included. We investigated the influence on homocysteine levels of a 3-month treatment with a daily oral dose of 4 mg 17beta-estradiol (ET) or 4 mg ET combined with 10 mg dydrogesterone (EPT); the comparison group received placebo treatment. We performed primed continuous infusions of L-[2H3-methyl-13C]methionine to assess steady-state flux rates of transmethylation, remethylation, and transsulfuration. Homocysteine concentration relationships with S-adenosylmethionine, S-adenosylhomocysteine, creatinine, albumin, vitamins B6 and B12, and folate status were determined as well. The mean change from baseline in homocysteine concentration by both treatments compared with placebo (ET, -13%; EPT, -10%) was accompanied by a decrease in the concentration of vitamin B6 (ET, -25%; EPT, -38%) and albumin (ET, -7%; EPT, -11%). No significant changes in flux rates were observed. In a .multiltivariate analysis, changes in homocysteine concentration were related to changes in albumin concentration. No relation to other variables was observed. We conclude that the ET- and EPT-induced homocysteine changes in this study were not accompanied by a significant change in methionine-homocysteine flux rates and hypothesize that an estrogen-induced lowering of homocysteine levels is primarily part of a change in albumin metabolism.