Vitamin D metabolites alphacalcidol and calcitriol (D-hormones) have been investigated for two decades, but few and conflicting results are available from high-quality randomized controlled trials. Our objectives were to provide an evidence-based update quantitatively summarizing their efficacy on bone mineral density (BMD) and fracture rate. We performed a systematic research of any randomized controlled trial containing relevant data, peer review, data extraction and quality scoring blinded for authors and data sources, and comprehensive meta-analyses of the relevant data. Inclusion criteria were: randomized controlled study, calcitriol or alphacalcidol, BMD or fractures in healthy/osteopenic/osteoporotic patients exposed or not to corticosteroids (CS). Analyses were performed in a conservative fashion using professional dedicated softwares and stratified by outcome, target patients, study quality, and control-group type. Results were expressed as effect size (ES) for bone loss or relative risk (RR) for fracture while allocated to D-hormones vs control. Publication bias and robustness were investigated. Of the trials that were retrieved and subsequently reviewed, 17 papers fitted the inclusion criteria and were assessed. Quality scores ranged from 20 to 100%, the mean (standard deviation) being 72 (22)%. Calcitriol and alphacalcidol were found to have the same efficacy on all outcomes at p>0.13. We globally assessed D-hormones effects in preventing bone loss in patients not exposed to CS, and found positive effect: ES=0.39 (p<0.001). For lumbar spine, this particular effect was 0.43 (p<0.001). D-hormones significantly reduced the overall fracture rates: RR=0.52 (0.46; 0.59) and both vertebral and non-vertebral fractures: RR=0.53 (0.47; 0.60) and RR=0.34 (0.16; 0.71), respectively. No statistical difference in response was observed between results from studies on healthy and osteoporotic patients or depending on the fact that controls were allowed to calcium supplementation. Treatment with D-hormones was evaluated for maintaining spinal bone mass in five trials of patients with CS-induced osteoporosis, and provided ES=0.43 at p<0.001. Only two studies specifically addressed the effects of calcitriol on spinal fracture rate. None of them provided significant results, and the global RR did not reach the significance level as well: RR=0.33 (0.07; 1.51). Our data demonstrated efficacy for DH on bone loss and fracture prevention in patients not exposed to CS and on bone loss in patients exposed to CS, in the light of the most reliable scientific evidence. Their efficacy in reducing the number of fractures in patients exposed to CS remains to be determined.
