Article Text

Download PDFPDF

Steroids cause osteoporosis
  1. M-M Gordon1,
  2. S Stevenson1,
  3. J A Hunter1
  1. 1Department of Rheumatology (Floor 7), Gartnavel General Hospital, 1053 Great Western Road, Glasgow G12 0YN, UK
  1. Correspondence to:
    Dr M-M Gordon;
  1. B Gudbjornsson2,
  2. U I Juliusson3,
  3. F V Gudjonsson4
  1. 2Centre for Rheumatology Research, University Hospital, Reykajvik, Iceland and Akureyri Central Hospitali, Akureyri, Iceland
  2. 3Akureyri Central Hospitali, Akureyri, Iceland
  3. 4Health Care Centre, Akureyri, Iceland

    Statistics from

    We read with great interest the article by Gudbjornsson and colleagues and concur with the accompanying leader by Dr Paget on the issue of corticosteroid associated osteoporosis.1,2 Osteoporosis is a major public health problem, associated with significant morbidity and mortality, and is estimated to cost £614 million annually in England and Wales alone. Despite well published guidelines on the prevention and treatment of corticosteroid associated osteoporosis, as a profession, we are failing to meet the targets set by these guidelines.3,4

    In the light of the American College of Rheumatology guidelines in 2001, we performed an audit of our current practice relating to the issue of steroid prescription, calcium supplementation, measurement of bone density, and the prescription of antiresorptive treatment to see if we had been adhering to the recommendations of the National Osteoporosis Society. Our rheumatology department has a continually updated database on all current and past patients who have attended our unit. This contains information on patient demographics, primary rheumatological diagnosis, comorbid conditions, current drug treatment, past disease modifying treatment (including corticosteroids), and records all patient generated events, including outpatient and inpatient episodes. From our database of over 10 000 patients, we identified 258 patients who were currently receiving prednisolone and had been taking the drug for a minimum of three months.

    Forty patients were selected at random and case records were then reviewed. We recorded information on patient demographics, current prednisolone dose, the maximum prednisolone dose, and the reason for the prescription of corticosteroids. We looked at other identifiable risk factors for osteoporosis, the co-prescription of calcium supplements and treatment for osteoporosis and whether the patients had ever had a bone density measurement. Finally, we telephoned a selection of the patients to inquire if they had ever received verbal or written information on osteoporosis.

    Forty patients (29 female) were randomly selected. The median patient age was 63 years (range 33–85). The patients were taking a mean daily dose of 6.7 mg prednisolone (range 1–45) and had been prescribed prednisolone for a median of 6 years (range 3 months–20 years). The most common reason for the prescription of prednisolone was for polymyalgia rheumatica for 14 (35%) of those selected, followed by systemic lupus erythematosus (SLE) in nine (23%), rheumatoid arthritis or associated complications for four (10%), and mixed connective tissue disease for three (8%). There were also isolated prescriptions for juvenile idiopathic arthritis, dermatomyositis, polymyositis, psoriatic arthritis, Wegener's granulomatosis, iritis, and unspecified systemic vasculitis.

    Encouragingly, 34 (85%) of our cohort were receiving some form of bone protective treatment: 22 (55%) were taking an oral bisphosphonate and one patient received intravenous pamidronate. Four of the postmenopausal women were taking hormone replacement therapy and two patients were receiving calcitriol. Twenty (50%) were prescribed calcium and vitamin D supplements, and this was the only treatment in eight (20%) of the cohort. However, of the six patients not receiving any form of bone therapy, five were over the age of 65 years and the one patient under the age of 65 years has been treated with prednisolone continuously for 20 years for SLE, up to a maximum dose of 80 mg/day.

    Twenty four (60%) of the 40 patients had bone density measured by dual x ray absorptiometry (DXA) scan. Of these, seven (29%) were normal, eight (33%) showed osteopenia, and nine (38%) demonstrated osteoporosis at the lumbar spine or the neck of the femur, or both. All patients who had either osteopenia or osteoporosis on DXA scan were treated with bone protective agents.

    We interviewed 24 patients by telephone. Eighteen (75%) recalled being informed of steroid side effects. Seven had received written information on steroids, but only three had received written information on osteoporosis.

    Although our results are encouraging, a significant number of patients are not being treated to prevent osteoporosis and reduce future fracture risk. The fact that all patients with an abnormal bone density scan are treated is reassuring, but we cannot be sure what proportion of patients who have not been scanned require treatment. In 1996, Walsh and colleagues found that only 14% of patients receiving long term prednisolone were receiving some form of preventive treatment against osteoporosis.5 Although matters have improved to a degree, a substantial proportion of patients treated with steroids is still undertreated and considerable progress has to be made nationally and internationally to prevent further bone associated morbidity among patients treated with corticosteroids.


    Authors' response

    We appreciate the comment by Dr Gordon and her coworkers on our article on the prevalence of decision making against steroid-induced osteoporosis, which was recently published in the Annals1 and which included a leader from Dr Paget.2 In their letter, they further highlight the importance of prevention against corticosteroid-induced osteoporosis. They also reported their experience at their rheumatological clinic with more than 10 000 patients. Surprisingly, only 2.6% of their patients with various rheumatological disorders were receiving long term treatment with corticosteroids, in comparison with 0.7% of our unselected population based cohort. More than half of their patients were receiving bisphosphonate, but unfortunately they did not report whether this was primary or secondary prevention or treatment against manifest osteoporosis, nor did they report the prevalence of fragility fractures in their patient group. Although this is a much higher proportion than we1 and others have previously reported,3,4 still 20% of their patients were not treated with any antiresorptive agent and 15% were neither receiving specific bone protective treatment nor calcium or vitamin D supplementation.

    These figures show that even in a specialist clinic with attention to osteoporosis, further work needs to be done in primary prevention against steroid-induced osteoporosis. Since we performed our study in northeast Iceland an Osteoporosis Clinic with DXA has been established in the study area, and the Directorate of Public Health in Iceland has published clinical guidelines concerning this issue.5 Thus, it will be of interest to re-perform our study in the near future for evaluation of the actual improvement in preventing bone morbidity in patients in need of long term treatment with corticosteroids.


    View Abstract

    Request Permissions

    If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.