Steroids and longterm use: bone loss can be prevented with calcium and vitamin D.

Clinical bottom line (level 1a)

  1. Patients on long term steroids who are given calcium plus vitamin D have less bone mineral loss in their lumbar spine and forearm, than those given calcium alone or placebo.
  2. The effect on non-traumatic fractures and side effects is still unclear.
Homik et al: Cochrane Library, Issue . Oxford: Update Software 1999; 3: -
Expires May 2001

The study

Systematic review of controlled trials of
  • Patients: on systemic corticosteroids
  • Intervention: calcium plus vitamin D compared with calcium alone or placebo
  • Outcome: bone mineral density, new non-traumatic fractures, adverse effects


    • Articles found in any using EMBASE and MEDLINE, 1966 to 1996 (search terms: bone diseases, osteoporosis, anti-inflammatory agents; steroidal, corticosteroid ) and hand searching of abstracts from various scientific meetings and reference lists of selected trials

    Selection criteria: as above
    Appraisal criteria: two investigators- detailed in text
    Articles excluded if: bone biopsy data only

    Five placebo-controlled randomised trials were identified. Four were double blinded. Patients (mainly with rheumatic arthritis) received a mean dose of 5.6 to 18.9 mg of prednisolone and were followed for a year. Patients received cholecalciferol (two studies) or a more active vitamin D metabolite (three studies).

    The evidence

    Outcome Time to outcome CER OR
    (95% CI)    		 NNT
    (95% CI)
    new non-traumatic fractures 12 months 5/39
    (13%)    		 0.6
    (0.1 to 2.4)     		-21
    (NNT = 7 to infinity;
    NNH = 9 to infinity)
    drop out due to adverse effects 12 months 4/63
    (6.4%)    		 1.9
    (0.5 to 6.4)     		20
    (NNT = 4 to infinity;
    NNH = 32 to infinity)

    • Bone mineral density of the lumbar spine had a mean difference in favour of those taking calcium and vitamin D of 2.6 g/cm ² (95% CI: 0.7 to 4.5)
    • Bone mineral density of the distal radius in favour of experimental patients was 2.5 g/cm ² (95% CI: 0.6 to 4.4).
    • Bone mineral density of the femoral neck had a mean difference of 0.4 g/cm ² (95% CI: -1.1 to 1.8).

    Comments

    1. Only two trials studied non-traumatic fractures and bone mineral density at the femoral neck, and follow-up was too short to demonstrate any clear effect. However, noting increase in bone mineral density at other sites, a reduction in fractures would be expected.
    2. The available studies did not attempt to determine whether Vitamin D (cholecalciferol) or di-hydroxy vitamin D (calcitriol) was superior.

    Citation

    1. Homik J, Suarez-Almacor ME, Shea B, et al: Calcium and vitamin D for corticosteroid-induced osteoporosis (Cochran Review). Cochrane Library, Issue . Oxford: Update Software 1999; 3: -
    Contributor: Chris Ball and Clare Wotton, April 2000
    Reviewer: Daniel Sontheimer

    Clinical Question.
    Patient on systemic corticosteroids
    Intervention or Exposure calcium plus vitamin D
    Comparison calcium alone or placebo
    Outcome fractures