A prospective, randomised, double-blind, placebo controlled primary prevention trial was undertaken in 28 patients commencing low to moderate doses of corticosteroids for the first time. Patients were randomised to intermittent cyclical etidronate (400 mg daily for 2 weeks) and calcium (500 mg daily for 11 weeks) or intermittent cyclical placebo with calcium. After 52 weeks of treatment, lumbar spine BMD increased by 1.8% in the etidronate group, while it decreased by 3.7% in the placebo group. The differences in bone loss rate were statistically significant (p<0.01) at both 6 and 12 months. Similar trends were observed at the proximal femur, but differences were not statistically significant. These results suggest that intermittent cyclical etidronate therapy is effective in the primary prevention of corticosteroid-induced bone loss at the lumbar spine.