The aim of this review with meta-analysis was to determine if there is a rationale to use activated forms of vitamin D3 to treat or prevent glucocorticoid-induced osteoporosis, and to compare the effect of active vitamin D3 metabolites with that of other anti-osteoporosis therapies. We performed a systemic search using MEDLINE/PubMed (1966-2003). Animal studies and clinical trials involving humans with data on therapy to treat or prevent glucocorticoid-induced osteoporosis with active vitamin D3 analogues were included. Animal studies and basic research studies with active vitamin D3 were reviewed (qualitative review). Meta-analysis (quantitative review) on clinical trials (including organ transplantation studies) was performed with percent change in lumbar spine bone mineral density or bone mineral content as the primary outcome measure; the secondary outcome measure was incidence of vertebral fractures. Fifty-four articles were found. Animal and basic research studies showed that active vitamin D3 analogues can inhibit bone loss during treatment with glucocorticoids. Concerning the effect on bone mineral density, the pooled effect size of active vitamin D3 analogues compared with no treatment, placebo, plain vitamin D3 and/or calcium was 0.35 (95% confidence interval (CI) 0.18, 0.52). Compared with bisphosphonates, the pooled effect size was -1.03 (95% CI -1.71, -0.36). The pooled estimate of the relative risk for vertebral fractures of active vitamin D3 analogues compared with no treatment, placebo, plain vitamin D3 and/or calcium was 0.56 (95% CI 0.34, 0.92) and compared with bisphosphonates it was 1.20 (95% CI 0.32, 4.55). Active vitamin D3 analogues not only preserve bone during glucocorticoid therapy more effectively than no treatment, placebo, plain vitamin D3 and/or calcium, but are also more effective in decreasing the risk of vertebral fractures. Bisphosphonates, however, are more effective in preserving bone and decreasing the risk of vertebral fractures than active vitamin D3 analogues.