Following a 52-wk randomized controlled trial of intermittent cyclic etidronate therapy in patients using corticosteroids, we performed a 52-wk open-label trial of calcium alone in 114 corticosteroid-treated patients to determine whether the beneficial effect of etidronate is maintained after the drug is discontinued. All patients were given 500 mg/d of elemental calcium. Sixty-one and 53 patients made up the former placebo and etidronate groups, respectively. A total of 89 (98%) of patients in the former placebo and etidronate groups remained on corticosteroids throughout the second year. The mean (SE) percentage change in bone mineral density of the lumbar spine, femoral neck, and trochanter were compared between groups. The difference between groups in mean percentage change from baseline (wk 0, initiation of etidronate or placebo therapy) in the bone density of the lumbar spine, femoral neck, and trochanter, following 104 wk, was 3.8 (0.9), 3.0 (1.1), and 4.3 (1.1), respectively (p < 0.05, all sites), in favor of the former etidronate group. While not significant, the former placebo group demonstrated a slightly larger rate of decline in bone density over the second year than the former etidronate group at all three sites. Following the discontinuation of etidronate therapy, there was no accelerated bone loss and there was evidence of a residual protective effect in both the lumbar spine and femoral neck for up to 1 yr posttreatment.