The aim of this study was to compare the metabolic effects of two presentations of 17 beta-oestradiol (E2) which are of recognized effectiveness in the prevention of post-menopausal bone loss, one being administered via the oral and the other via the percutaneous route. During this prospective, randomized study, 32 patients were treated for 2 mth with either 2 mg/day of oral micronized E2 (n = 16) or 1.5-3 mg/day of percutaneous E2 (n = 16). Both regimens proved efficacious, since significant increases in oestrone (E1) and E2 concentrations ranging up to mid-follicular values were observed. In the percutaneous-treatment group we noted a significant decrease in triglycerides (TG), without any significant changes in high-density lipoprotein cholesterol (HDL-C) or low-density lipoprotein cholesterol (LDL-C). In the oral-treatment group, we saw no significant increase in HDL-C, although significant increases were observed in body weight, TG, plasma renin substrate (PRS) and sex-hormone-binding globulin (SHBG) as well as significant decreases in antithrombin III (AT III) activity and antigen. All of these metabolic variations led us to the conclusion that oral E2 at the dose established as effective in preventing post-menopausal osteoporosis may, even when micronized, alter certain metabolic and haemostatic parameters in a population characterized by increases in cardiovascular risk factors and morbidity. Oral oestrogen replacement therapy should therefore continue to be used only in carefully selected patients and be strictly followed up by systematic checks on a series of metabolic criteria.