Both trabecular and cortical bone mineral content reduction physiologically occur during the premenopausal period and, overall, at the time of menopause. It is related to plasma estrogen fall. Several studies have examined the relationship between bone density and oral contraceptives (OCs), but their results disagree. The different results may be due to the small number of patients investigated, the different OC formulations, and to the multifactorial nature of the bone loss phenomenon. Furthermore, the use of different techniques and different sites for measuring the bone density may limit the interpretation and reproducibility of the reported findings. Our previous studies evidenced that: (1) in premenopausal (40-45 years) women (n = 26), OCs, over a period of 2 years, do not induce any change either in bone mineral density or in bone metabolism biochemical markers (osteocalcin, alkaline phosphatase, urinary hydroxyproline/creatinine ratio; and (2) independent from the therapy length, the premenopausal use of OCs does not modify the bone mineral loss in postmenopausal patients (n = 40). The present study compared, over a period of 36 months, in perimenopausal (46-53 years; follicle stimulating hormone < or = 35 mU/ml; estradiol < 50 pg/ml) women, the effect of OCs (n = 8) on bone metabolism. The results were compared with those of a matched control group (n = 9). In the treated group, a slight, though not significant, bone mass improvement was observed during the entire therapy period, whereas, in the control group, a continued and progressive bone mass loss was observed and a significant decrease was shown at the end of the observational period. Our data suggest that OCs have no significant effects on bone metabolism in women during their childbearing years. However, OCs may be beneficial in inhibiting bone reabsorption in pre- and postmenopause. Furthermore, the pill, also in virtue of its contraceptive effect, would be the first presidium in hormone therapy in the premenopause.