Corticosteroid treatment diminishes bone mass and alters bone quality. The objective was to evaluate bone in corticosteroid-treated patients and controls and in fractured and non-fractured patients treated with corticosteroids using both X-ray densitometry (DEXA) and ultrasound. We evaluated 34 women aged 58 ± 14 years (X ± SD), who had been on long-term low dose prednisone therapy for at least 6 months, and who had never received specific treatment for osteoporosis. Bone mineral density of total skeleton (TS), lumbar spine (LS), femoral neck (FN), and vertebral morphometry (MXA) were measured by DEXA. Speed of sound (SOS), broadband ultrasound attenuation (BUA) and stiffness were measured using an Achilles Plus system. Forty-two healthy women served as controls. Both densitometric and ultrasound parameters in the patients were significantly diminished compared with controls: TS: P < 0.002, LS: P < 0.025, FS: P < 0.005, Stiffness: P < 0.001, BUA: P < 0.002 and SOS: P < 0.002. The percentage of patients with a Z score below -2 was higher in Stiffness and BUA: 38% and 47%, respectively, compared with a range of 16-24% in the other parameters (P < 0.05 BUA vs. DEXA measurements). Eleven patients with previous bone fracture had values lower than the non-fractured patients, both according to DEXA and ultrasound measurements, but the difference was only significant for BUA (P < 0.02). BUA of the calcaneus was more effective in detecting the specific skeletal alterations and fracture risk of the group of patients receiving chronic corticosteroid treatment.