Long-term effects of high doses of inhaled corticosteroids (ICS) on bone density and metabolism are still uncertain. Fifty-one patients (37 male, 14 female) using beclomethasone or budesonide at a daily dose > 800 microgram/d (high-dose group [Group HD] mean: 983 microgram/d [prescribed dose x estimated compliance]) or no or < 500 microgram/d (control group [Group C] mean: 309 microgram/d) for more than 5 yr were enrolled in this study. Each had, 3 yr ago and at this last evaluation, a clinical evaluation and measurements of expiratory flows and of bone density and bone metabolism markers. Lumbar spine bone density (last visit) was similar in the two groups with respective values of 0.94 +/- 0.03 (HD) and 0.96 +/- 0.03 g/cm2 (C) (p > 0.05). T and Z scores were -1.21 +/- 0.19 and -0.70 +/- 0.18 (HD), -0.95 +/- 0.25 and -0.47 +/- 0.21 (C) respectively (p > 0.05). A correlation was found between the decrease in bone density and the mean daily dose of corticosteroid in Group HD although these changes were quite small, mean bone density being unchanged over the 3-yr period. Serum and urinary parameters were similar in the two groups. Furthermore, neither initial bone density nor any of the biological parameters could predict changes in bone density over a period of 3 yr. In conclusion, bone density was similar in both study groups and not significantly different over a 3-yr period. Neither initial bone density nor biological markers of bone metabolism helped to predict changes in bone mass.