Thirty-eight RCTs involving more than 57,308 participants (n not reported for some trials) met the inclusion criteria. However, the results of several RCTs that did not meet the inclusion criteria were also reported, as were 5 meta-analyses and 8 observational studies.
Only the results from full publications of RCTs that met the stated inclusion criteria are reported here (except where the interventions were only reported in conference abstracts).
Bisphosphonates, alendronic acid.
One trial showed a reduction in the incidence of women with new radiographic or symptomatic vertebral fractures, hip or any symptomatic fracture at 3 years in women who had at least one radiographic vertebral fracture at baseline. In the same trial, alendronic acid was also associated with a reduction in fracture risk in women with low bone mineral density (BMD) or established osteoporosis. Two more RCTs were also reported to show significant reductions in the incidence of women with new radiographic vertebral fractures and in the cumulative incidence of symptomatic nonvertebral fractures. The trials were reported to be of a high quality.
Bisphosphonates, risedronic acid plus calcium.
Two large RCTs reported statistically-significant reductions in the risk of radiographic vertebral fractures in women with either 2 or more radiographic vertebral fractures at baseline or one radiographic vertebral fracture and low BMD (T score of less than -2.0).
Bisphosphonates, etidronic acid.
Two small trials reported significant reductions in radiographic vertebral fracture incidence during 2 to 3 years' treatment. A small 4-year RCT of etidronic acid alone and in combination with hormone replacement therapy (HRT) showed no significant reduction in radiographic vertebral fracture risk.
Bisphosphonates, pamidronic acid.
One very small RCT (n=48) showed no difference in the incidence of fractures or height loss.
Bisphosphonates, tiludronic acid plus calcium.
Four RCTs showed no statistically-significant effect of intermittent cyclical tiludronic acid (50 or 200 mg) on the incidence of radiographic vertebral fractures, compared with placebo.
Bisphosphonates, clodronic acid.
An interim analysis of 1-year data from a planned 3-year trial showed no statistically-significant difference between clodronic acid and placebo, with regard to the incidence of new radiographic vertebral fractures.
Oestrogen replacement therapy.
One RCT showed no significant reduction in incidence of symptomatic fractures; however, the effect on radiographic vertebral fractures could not be determined. Another RCT found that the incidence of nonspine fracture in the HRT plus vitamin D group was significantly lower than in the placebo group (RR 0.29, 95% CI: 0.10, 0.90). However, approximately one third of the participants dropped out of the HRT groups in this trial; this was three times the rate in the placebo or vitamin D groups. One small RCT showed no significant difference between the groups (RR 0.6, p>0.17). Another small RCT also showed no significant difference between the groups.
SERMs, raloxifene.
One large RCT reported a significant reduction in the incidence of radiographic vertebral fractures with raloxifene in 60 mg (RR 0.7, 95% CI: 0.5, 0.8) and 120 mg doses (RR 0.6, 95% CI: 0.4, 0.7), and in the incidence of clinically diagnosed vertebral fractures (RR 0.4, 95% CI: 0.3, 0.7, both groups pooled). No significant difference between the groups was found in the incidence of women with nonvertebral fractures (RR 0.9, 95% CI: 0.8, 1.1) or hip fractures (RR 1.1, 95% CI: 0.6, 1.9). A much smaller RCT found no significant effect of raloxifene on the incidence of either nonvertebral fractures or radiographic vertebral fractures.
SERMs, tamoxifen.
A large RCT designed to assess the incidence of breast cancer also reported outcomes for hip, wrist and symptomatic vertebral fractures. No significant differences were found.
Calcitonin.
One small trial found a significant reduction in the incidence of both radiographic vertebral fractures and peripheral fractures. However, there were several methodological limitations to this trial.
Calcium plus vitamin D.
One large study reported that calcium (1,200 mg daily) plus vitamin D (800 IU daily) significantly reduced the incidence of hip (RR 0.73, 95% CI: 0.23, 0.99) and nonvertebral fractures (RR 0.72, 95% CI: 0.60, 0.84) in a population with a high prevalence of vitamin D deficiency and a low calcium intake.
Calcium.
One RCT of calcium (1,200 mg/day) found no significant differences from placebo.
Vitamin D.
Two RCTs found no effect of vitamin D on the incidence of hip or nonvertebral fractures. Three small RCTs found no effect of calcitriol on vertebral fracture risk or nonvertebral fractures. Two small trials of alfacalcidol found no significant reductions in fracture risk. A third trial did report a significant reduction in vertebral fracture risk, but the trial was of only 6 months' duration.