A group of 366 healthy, white postmenopausal women, aged 50-81 years, mean age 66 years, were selected from the screened population of Scandinavians who were part of a multicenter study of the efficacy of tiludronate, a new bisphosphonate, in established postmenopausal osteoporosis, Eighty-eight women had a lumbar spine bone mineral density (BMD) above 0.860 g/cm(2), and 278 women had a BMD below 0.860 g/cm2. Spinal fracture was diagnosed from lateral spine X-ray studies and defined as at least 20% height reduction (wedge, compression, or endplate fracture) in at least one vertebra (T4-L4), Bone resorption was assessed by measurement of the urinary excretion of type I collagen degradation products by the CrossLapsTM enzyme-linked immunoassay (ELISA). Bone formation was assessed by ELISA measurement of the N-terminal-midfragment as well as the intact serum osteocalcin (OCN-MID), thus omitting the influence of the instability of osteocalcin caused by the labile 6 amino acid C-terminal sequence. The women were divided into groups with high or low bone turnover according to the concentrations of urinary CrossLapsTM or OCN-MID. Women in the quartiles with the highest concentrations of CrossLaps [519 +/- 119 mu g/mmol (SD)] or OCN-MID [44.6 +/- 7.5 ng/ml (SD)] had 10-16% lower spinal BMD compared with women in the lowest quartiles (CrossLaps 170 +/- 48 mu g/mmol (SD), and OCN-MID [22.1 +/- 3.0 ng/ml (SD)] (P < 0.0004). The prevalences of spinal fracture were 25 to 29% in the lowest quartiles, whereas the prevalences in the highest quartiles were almost double-53-54% (P < 0.006). If the women were subgrouped according to spinal BMD and prevalence of spinal fracture, corresponding results were found. Women with a BMD less than 0.860 g/cm(2), without or with spinal fracture (n = 136 and n = 142), had 36-43% higher concentration of CrossLaps (P = 0.0001) and 11-15% higher concentration of OCN-MID (P < 0.02), as compared with women with a BMD above 0.860 g/cm(2) and no spinal fracture (n = 84). In conclusion, the results indicate a strong association among high bone turnover, low bone mass, and prevalence of spinal fracture, which supports the theory that high bone turnover is a risk factor for spinal fracture and osteoporosis.