Objective: To evaluate the effects of a 6 month administration of raloxifene hydrochloride, a selective estrogen receptor modulator which was recently approved for the prevention of osteoporosis, on serum gonadotropin and prolactin (PRL) levels and on TRH-stimulated PRL responsiveness in postmenopausal women who have not undergone estrogen replacement therapy. Design and methods: Sixteen healthy postmenopausal women were divided into two groups on the basis of their bone status, evaluated by dual energy X-ray absorptiometry at the lumbar level. Eight women (chronological age 52:4^4:1 (S.D.) years, menopausal age 42:4^3:9 years), in whom T-score L2 -L4 was less than 2 2.5 S.D., were treated with raloxifene (60 mg p.o.) administered daily for 6 months (group 1), while the other eight women (chronological age 52:6^2:5 years, menopausal age 42:13 :6 years), in whom the T-score L2 -L4 ranged between 21 and 22.5 S.D., were used as a control group (group 2). Serum PRL, FSH, LH and 17b-estradiol (E2) levels were evaluated at baseline and after 3 and 6 months of treatment. In all subjects, PRL responsiveness to TRH (200 mg i.v.) administration was evaluated at baseline and at the end of the study. Results: At baseline, mean PRL, LH and FSH levels were not significantly different in the two groups (PRL 133:6^21:7 vs 136:7^28:1 mIU=l (NS), LH 25:1^6:8 vs 24:4^6:7 mIU=ml (NS), FSH 74:42 5:0 vs 71:1^24:1 mIU=ml (NS), in group 1 and group 2 respectively). No significant variations in serum FSH and LH values, in either group, or in serum PRL levels in group 2, were observed at the 3 and 6 month examinations. On the contrary, serum PRL values decreased significantly in group 1 after 3 months ð100:1^47:7 mIU=l; P , 0:05Þ and 6 months ð81:5^30:2 mIU=l; P , 0:001Þ: At baseline, no significant differences were observed in the TRH-stimulated serum PRL peak between the groups (1015:4^30:5 vs 1030:2^25:7 mIU=l in group 1 and in group 2 respectively), while it decreased significantly at the 6 month examination in group 1 ð770:5^47:4 mIU=l; P , 0:001Þ and it was significantly lower than in group 2 ð1068:1^301:8 mIU=l; P ¼ 0:02Þ: Serum E2 was not detected at baseline and at each examination, in all patients. Conclusions: The decrease of PRL values induced by long-term raloxifene administration in postmenopausal women could be explained by a direct antiestrogenic effect of raloxifene on lactotrope cells or by the recently suggested increase of opiatergic tone on the hypothalamic -pituitary region.