We and others have demonstrated that substantial loss of bone mineral density (BMD) is very common following alloBMT, in part due to prolonged use of glucocorticoids. Intravenous (IV) pamidronate is a potent bisphosphonate with efficacy in preventing glucocorticoid-induced bone loss in the non-BMT setting. 116 alloBMT recipients were randomised at 5 institutions to receive pamidronate 90mg IV monthly from day-7 to one year post-transplant (n=63) or no pamidronate (n=53) in an open label, prospective, controlled trial. All patients received oral vitamin D and calcium supplements and all women also received hormone therapy with an oestrogen and progestin. The primary end-point was the reduction in bone loss from the lumbar spine, femoral neck and total hip at 12 months post BMT. Age, sex and conditioning regimen (total body irradiation versus chemotherapy only) were not significantly different between the groups. 37 patients were not evaluable, predominantly due to early death (n=29) or protocol violations (n=4). Significant reductions in BMD loss at 12 months were seen in all 3 evaluated sites in patients treated with pamidronate.

SiteNo pamidronatePamidronatep value
n% change*n% change*
*percentage change in BMD at 12 months 
Lumbar Spine 28 −3.77 46 2.53 <.0001 
Femoral neck 27 −9.33 45 2.82 <.0001 
Total Hip 23 −8.35 38 −3.32 .0072 
SiteNo pamidronatePamidronatep value
n% change*n% change*
*percentage change in BMD at 12 months 
Lumbar Spine 28 −3.77 46 2.53 <.0001 
Femoral neck 27 −9.33 45 2.82 <.0001 
Total Hip 23 −8.35 38 −3.32 .0072 

There were no significant differences in the 12-month changes associated with age class (<30, 30–40, 40–50, >50 years) or sex. In preliminary analyses of steroid dose subgroups, statistically significant improvements in BMD loss were found in patients whose average daily equivalent prednisolone dosage in the first six months post transplant was i) >25mg for all sites and ii) 10–25mg daily for the lumbar spine. We conclude that prophylactic pamidronate significantly reduces bone loss from the spine and hip after alloBMT. The impact on clinically relevant endpoints such as the subsequent incidence of fractures and avascular necrosis in these patients will determine the utility of this intervention.

Author notes

Corresponding author

Sign in via your Institution