Although fluoride salts have been shown to be capable of linearly increasing spinal bone mineral density (BMD) in postmenopausal osteoporosis, the effects of this gain in density on the vertebral fracture rate remain controversial. We conducted a 2-year multicenter, prospective, randomized, double-masked clinical trial in 354 osteoporotic women with vertebral fractures (mean age 65.7 years). They received either fluoride (208 patients), given as sodium fluoride (50 mg/day) or as monofluorophosphate (200 mg/day or 150 mg/day), or a placebo (146 patients). All patients received daily supplements of 1 g of calcium (Ca) and 800 IU of vitamin D2 (D). A 1-year open follow-up on Ca-D was obtained in 124 patients. After 2 years the fluoride group and the Ca-D group had increased their lumbar BMD by 10.8% and 2.4% respectively (p = 0.0001). However, the rate of patients with at least one new vertebral fracture, defined by semiquantitative assessment and evaluable on an intention-to-treat basis in 89% of patients, was similar in the fluoride groups and the Ca-D group. No difference between the three fluoride regimens was found. The percentage of patients with nonvertebral fractures was not different in the fluoride and Ca-D groups (1.9% and 1.4% respectively for hip fractures). A lower limb pain syndrome occurred more frequently in the fluoride groups. In the 124 patients followed for 1 year after cessation of fluoride therapy, the percentage of patients with at least one new vertebral fracture after 36 months was identical to the percentages in the previous fluoride group and the Ca-D group. We conclude that fluoride-Ca-D regimen was no more effective that Ca-D supplements for the prevention of new vertebral fractures in women with postmenopausal osteoporosis.