Postmenopausal estrogen-replacement therapy may reduce the risk of cardiovascular disease, and this beneficial effect may be mediated in part by favorable changes in plasma lipid levels. However, the effects on plasma lipoprotein levels of postmenopausal estrogens in the low doses currently used have not been precisely quantified, and the mechanism of these effects is unknown. We conducted two randomized, double-blind crossover studies in healthy postmenopausal women who had normal lipid values at base line. In study 1, 31 women received placebo and conjugated estrogens at two doses (0.625 mg and 1.25 mg per day), each treatment for three months. In study 2, nine women received placebo, oral micronized estradiol (2 mg per day), and transdermal estradiol (0.1 mg twice a week), each treatment for six weeks. The metabolism of very-low-density lipoprotein (VLDL) and low-density lipoprotein (LDL) was measured by endogenously labeling their protein component, apolipoprotein B. In study 1, the conjugated estrogens at doses of 0.625 mg per day and 1.25 mg per day decreased the mean LDL cholesterol level by 15 percent (95 percent confidence interval, 11 to 19 percent; P less than 0.0001) and 19 percent (95 percent confidence interval, 15 to 23 percent; P less than 0.0001), respectively; increased the HDL cholesterol level by 16 percent (95 percent confidence interval, 12 to 20 percent; P less than 0.0001) and 18 percent (95 percent confidence interval, 14 to 22 percent; P less than 0.0001), respectively; and increased VLDL triglyceride levels by 24 percent (95 percent confidence interval, 8 to 40 percent; P less than 0.003) and 42 percent (95 percent confidence interval, 26 to 58 percent; P less than 0.0001), respectively. In study 2, oral estradiol increased the mean concentration of large VLDL apolipoprotein B by 30 +/- 10 percent (P = 0.05) by increasing its production rate by 82 +/- 18 percent (P less than 0.01). Most of this additional large VLDL was cleared directly from the circulation and was not converted to small VLDL or LDL. Oral estradiol reduced LDL cholesterol concentrations by 14 +/- 3 percent (P less than 0.005), because LDL catabolism increased by 36 +/- 7 percent (P less than 0.005). The oral estradiol increased the HDL cholesterol level by 15 +/- 2 percent (P less than 0.0001). Transdermal estradiol had no effect. The postmenopausal use of oral estrogens in low doses favorably alters LDL and HDL levels that may protect women against atherosclerosis, while minimizing potentially adverse effects on triglyceride levels. The decrease in LDL levels results from accelerated LDL catabolism; the increase in triglyceride levels results from increased production of large, triglyceride-rich VLDL.