To characterize the pattern of biochemical markers of bone metabolism and femoral bone mineral density in eumenorrheic and oligomenorrheic perimenopausal women, and assess the effects of a low-dose oral contraceptive (OC) on bone metabolism and femoral bone density, bone biochemical markers and femoral bone density (measured at the neck, Ward's triangle and trochanter regions) were evaluated in a longitudinal 2-year follow-up study. The study was conducted in healthy, normally menstruating perimenopausal women (n = 18), perimenopausal oligomenorrheic women (n = 18), and perimenopausal oligomenorrheic women treated with an OC containing 20 microg ethinylestradiol plus 0.15 mg desogestrel (n = 19). The results were analyzed by factorial or repeated measures analysis of variance, as appropriate. During the observation period, in normally menstruating women there were no changes in the menstrual cycle, plasma FSH and estradiol levels, biochemical markers of bone turnover or femoral bone density. In oligomenorrheic untreated women an increase in cycle length with a concomitant decrease in plasma estradiol and an increase in plasma FSH levels were found (p < 0.05). In this group a significant increase in urinary excretion of hydroxyproline and in plasma osteocalcin levels with a concomitant significant decrease in femoral bone density (p < 0.05) occurred. In OC-treated women, osteocalcin plasma levels and urinary excretion of hydroxyproline significantly (p < 0.05) decreased, leading to a significant (p < 0.05) increase in femoral bone density. It is concluded that perimenopausal OC administration can avoid the increase in bone turnover and the decrease in femoral bone density due to the perimenopausal impairment of ovarian function.