Steroids: bisphosphonates reduces loss of lumbar bone mineral density.

Clinical bottom line (level 1a)

  1. Patients with inflammatory disorders on long-term steroids who take bisphosphonates compared with placebo lose less lumbar bone mineral density.
  2. There is no clear effect on loss of femoral bone mineral density or new vertebral fractures.
Homik et al: The Cochrane Library 1999; (3): -
Expires September 2004

The study

Systematic review of all controlled clinical trials of
  • Patients: adults with underlying inflammatory disorders taking systemic corticosteroids a mean steroid dose of 7.5 mg per day
  • Intervention: bisphosphonates within 3 months of starting steroids +/- calcium or vitamin D compared with placebo +/- calcium or vitamin D
  • Outcome: steroid-induced osteoporosis, change in bone mineral density in the lumbar spine or femoral neck at one year, new fractures, withdrawals due to adverse effects
Articles found in English (other languages found though not translated) using Cochrane Musculoskeletal Group, Medline, Embase, 1966 to 1997 (search terms: strategy detailed in text ) and and selected hand-searching of reference lists, scientific abstracts from relevant meetings in the last 5 years. Current Contents was searched for the previous 6 months.

Selection criteria: see above: randomised and quasi-randomised trials
Appraisal criteria: Jadad's criteria; by 2 independent reviewers
Articles excluded if:
  • children
  • on bisphosphonates within the last 6 months


13 trials involving 842 patients: mainly using etidronate. Four trials reported data on vertebral fractures, thirteen on lumbar bone mineral density and eight on femoral neck density
Studies were found to be heterogeneous for bone mineral density

The evidence

Outcome Time to outcome CER OR
(95% CI)		 NNT
(95% CI)
new vertebral fractures 12 months 21/160
(13.1%)		 0.76
(0.37 to 1.53) 		35
(NNT = 13 to infinity;
NNH = 18 to infinity)

  • for homogenous trials: bisphosphonates v. placebo: weighted mean difference
    • femoral bone mineral density at 12 months: 0.52 (95% CI: -0.014 to 1.05)
    • lumbar bone mineral density at 12 months: 4.1 (95% CI: 3.4 to 4.7)

Comments

  1. Only four studies reported on new vertebral fractures - therefore this systematic review cannot exclude any potential benefit from bisphosphonates.
  2. Three studies found an increased drop-out rate in the treatment group. The major adverse effect was nausea. However odds ratios could not be calculated since three other studies reported no dropouts in either group.

Citation

  1. Homik J, Cranney A, Shea B, et al: Bisphosphonates for steroid-induced osteoporosis (Cochrane Review): Oxford: Update Software. The Cochrane Library 1999; (3): -
Contributor: Chris Ball and Bob Phillips, September 1999
Reviewer:

Clinical Question.
Patient long term steroids
Intervention or Exposure bisphosphonates
Outcome fractures, osteoporosis