Clinical Endocrinology

Clinical Question: What is the best approach to managing glucocorticoid-induced osteoporosis?

Authors


Juliet Compston, Department of Medicine, Cambridge University Hospitals NHS Foundation Trust, Cambridge, UK. Tel.: +44 1223 336867; Fax: +44 1223 336846; E-mail: jec1001@cam.ac.uk

Summary

Glucocorticoid-induced osteoporosis is common, and the resulting fractures cause significant morbidity and mortality. Rapid bone loss and increased fracture risk occur soon after the initiation of glucocorticoid therapy and are dose dependent. The increase in fracture risk is partly independent of bone mineral density, probably as a result of changes in bone material properties and increased risk of falling. Fracture risk can be assessed using the FRAX® algorithm, although risk may be underestimated in patients taking higher doses of glucocorticoids. Because of the rapidity of bone loss and increase in fracture risk after the start of glucocorticoid therapy, primary prevention should be advised in high-risk individuals, for example older women and men, individuals with a previous fracture history and those with low bone mineral density. Bisphosphonates are the front-line choice for the prevention of fracture in the majority of glucocorticoid-treated patients, with teriparatide as a second-line option. Calcium and vitamin D supplements should be co-prescribed unless there is evidence of an adequate dietary calcium intake and vitamin D status.

Introduction

Osteoporosis is a common complication of glucocorticoid therapy but despite greater awareness in recent years it remains under-diagnosed and under-treated.1–3 Although similar in some respects to postmenopausal osteoporosis, glucocorticoid-induced osteoporosis has distinct characteristics that have important implications for its management in clinical practice. In particular, rapid bone loss occurs early after initiation of therapy and increases in fracture risk are seen within the first 6 months of treatment.4 These features emphasize the importance of early intervention with bone protective therapy in individuals at high risk of fracture.5

Pathogenesis of fractures in glucocorticoid-treated patients

Glucocorticoids increase fracture risk by a variety of mechanisms. Direct effects on bone include an early, transient increase in bone resorption accompanied by decreased bone formation that persists throughout the duration of glucocorticoid therapy. Effects on bone also occur as a result of hypogonadism, increased calcium loss from the intestine and kidney, and vitamin D insufficiency.5

Although the adverse effects of glucocorticoids on bone mineral density (BMD) are well documented, the increase in fracture risk that occurs is partially independent of BMD probably because of increased risk of falling, muscle weakness and frailty, and changes in bone material properties that are not captured by BMD measurements.6 In addition, the underlying disease for which therapy is given may contribute to bone loss as a result of inflammation, poor nutrition and immobility. The partial independence of glucocorticoid therapy from BMD as a risk factor for fracture is consistent with data indicating that fracture in glucocorticoid-treated individuals occurs at a higher BMD than in those not receiving glucocorticoids.7 Most available data indicate that, in individuals taking continuous oral glucocorticoids, bone loss and fracture risk increase with the dose and duration of therapy.8 The skeletal effects of inhaled and intermittent oral glucocorticoids are less well documented although there is some evidence that high doses of inhaled glucocorticoids may be associated with increased fracture risk.9–12

Assessment of fracture risk in the glucocorticoid-treated patient

Risk factors for fracture in glucocorticoid-treated individuals have not been as extensively studied as in untreated postmenopausal and older men. However, age, female gender, falls history, previous fracture, low body mass index, smoking, co-morbidities and other medications were found to be independently associated with fracture risk in a large study of glucocorticoid-treated individuals in the General Practice Research Database.13 Most of these risk factors are included in the WHO-supported fracture risk algorithm FRAX® (Sheffield, UK), which includes glucocorticoid therapy as an independent risk factor for fracture.

FRAX® is based on the principle that clinical risk factors that are partially independent of BMD can be used with or without BMD measurement to improve fracture risk prediction.14 The evidence base on which FRAX® was constructed is provided by a number of large, population-based, prospective cohort studies with wide geographical representation.15 However, there are some limitations associated with the use of FRAX®. First, the calculation in FRAX® of the increase in fracture risk associated with glucocorticoid therapy was based on both past and present glucocorticoid use, making underestimation of fracture risk likely in some individuals on current therapy. Secondly, dose and duration of glucocorticoids are not accounted for in the algorithm so that fracture risk is likely to be underestimated in those taking high doses. Similarly, there is no dose response in FRAX® for other risk factors such as tobacco use and previous fracture. Finally, the predictive value of FRAX® has mainly been documented for nonvertebral fractures whereas vertebral fractures are more common in patients receiving glucocorticoids, particularly in the early stages of treatment. These issues do not invalidate the use of FRAX® but emphasize the need for clinical judgment in the assessment of individual patients. It should be noted that FRAX® cannot be used in premenopausal women or men aged <40 years, is not designed to be used in individuals receiving bone protective treatment and uses femoral neck or total hip, but not spine, BMD.

Recent guidelines from the United States incorporate a FRAX®-based approach to fracture risk assessment of glucocorticoid-treated individuals, using selected FRAX® clinical risk factors and femoral neck BMD.16 Risk is classified as a low (<10%), middle (10–20%) or high (≥20%) 10-yr probability of a major osteoporotic fracture (hip, spine, wrist or humerus). Patients with a BMD T-score ≤−2·5 or with a previous fragility fracture are automatically included in the high-risk group. These risk categories are then used to determine treatment strategies, depending on the dose of glucocorticoids. European guidelines are currently being developed and are likely to adopt a similar, although not identical approach. The 2002 Royal College of Physicians Guidelines recommend primary prevention in men and women aged over 65 years and all individuals with a past history of fragility fracture, whilst secondary prevention should be considered in individuals with a BMD T-score ≤−1·5 and in those who suffer a fracture whilst taking glucocorticoid therapy.17

Investigations

In addition to measurement of BMD, imaging of the lumbar and thoracic spine using conventional radiographs or DXA-based imaging should be performed in high-risk patients, for example those with low spine BMD, height loss or spinal deformity, to exclude vertebral fractures. Other secondary causes of osteoporosis should be excluded in patients with low BMD and/or fractures and vitamin D status should be assessed by measurement of serum 25-hydroxyvitamin D levels.

Management of glucocorticoid-induced osteoporosis

General measures

Advice about lifestyle may be useful in reducing the adverse effects of glucocorticoids on bone, although there is no direct evidence that such measures reduce fracture risk. An adequate dietary intake of calcium should be recommended (around 750–1000 mg/day) and good nutrition and maintenance of a normal body weight should be encouraged. Vitamin D insufficiency, where present, should be corrected. Patients should be assessed for their risk of falling and, where appropriate, action should be taken to reduce falls risk. Physical activity should be advocated within the limits imposed by the underlying disease and smoking and alcohol abuse avoided. The dose of glucocorticoids should be constantly reviewed, with reduction in the dose or cessation of therapy where possible. In addition, the use of glucocorticoid-sparing agents or alternative formulations (e.g. topical or inhaled) should be considered.

Pharmacological interventions

A number of agents have been evaluated for the prevention and treatment of glucocorticoid-induced osteoporosis. However, evidence for their efficacy is based on BMD changes in bridging studies of interventions shown to reduce fracture in postmenopausal women with osteoporosis, and reduction in fracture risk has not been a primary end point of any study in glucocorticoid-treated patients. Nevertheless, some information about effects on vertebral fracture is available from secondary end points or safety data. At the present time, alendronate, etidronate, risedronate, zoledronate and teriparatide are widely approved for the management of glucocorticoid-induced osteoporosis (Table 1).

Table 1.   Approved pharmacological interventions for the management of GIOP
InterventionDosing regimenFormulationPopulation
  1. *Only the once daily regimens are approved for glucocorticoid-induced osteoporosis.

Alendronate5 or 10 mg once daily 70 mg once weekly*OralNot specified
Etidronate400 mg daily for 2 weeks every 3 monthsOralNot specified
Risedronate5 mg once daily 35 mg once weekly*OralPostmenopausal women
Zoledronate5 mg once yearlyIntravenous infusionMen and postmenopausal women
Teriparatide20 μg once dailySubcutaneous injectionMen and postmenopausal women

The majority of treatment studies have been short-term (approximately 1 year in duration) and have focused mainly on postmenopausal women. Although men and premenopausal women have been included in some studies, the numbers involved have been small and thus the evidence base for efficacy is weaker in these populations. In addition, some diseases, particularly inflammatory bowel disease, have been under-represented.

Because rapid bone loss and increased fracture risk occur soon after the initiation of glucocorticoid treatment, the timing of bone protective therapy is critical. Primary prevention, in which bone protective therapy is initiated at the same time as glucocorticoids, is therefore recommended in individuals at high risk of fracture, whereas secondary prevention may be appropriate for those who are at lower risk at the start of glucocorticoid therapy but progress to a higher risk category with time.

Bisphosphonates.  Bisphosphonates have been most widely evaluated and are generally regarded as the first-line option in the majority of patients. Alendronate, etidronate, risedronate and zoledronate are all approved for management of glucocorticoid-induced osteoporosis, although the wording of the indications differs somewhat between agents. All these drugs have been shown to prevent or reduce bone loss in the spine and hip, and subgroup analyses have shown reduction in risk of vertebral fracture for etidronate, alendronate and risedronate. Reduction in nonvertebral fractures, including hip fracture, has not been demonstrated probably because of the small number of these fractures in the clinical trials.18–27

The effects of zoledronate were evaluated in a randomized, double-blind, double-dummy noninferiority study.27 Zoledronate, 5 mg intravenously once yearly, was demonstrated to be noninferior (the primary end point) and in fact proved to be superior to risedronate in its effect on spine and hip BMD, both in the prevention subgroup (glucocorticoid treatment for <3 months) and in the treatment subgroup (glucocorticoid therapy for ≥3 months). The study was not powered to demonstrate differences in anti-fracture efficacy between the two agents. Because of its intravenous mode of administration, zoledronate is a useful option in individuals who cannot tolerate oral bisphosphonates, who have intestinal malabsorption or who are unlikely or unable to comply with oral bisphosphonate therapy.

Upper gastrointestinal side effects may occur with oral alendronate and risedronate, particularly if the dosing instructions are not closely followed. An acute phase reaction occurs in up to one-third of patients receiving their first intravenous infusion of zoledronate, although recurrence in subsequent infusions is rare. Osteonecrosis of the jaw and atypical fractures are very rare adverse events that may be causally linked to bisphosphonate therapy, and there is some evidence that glucocorticoid therapy is a risk factor for these conditions.28 Bisphosphonates cross the placenta and should be used with caution in premenopausal women and avoided in pregnancy.

Teriparatide.  Intermittent administration of parathyroid hormone (PTH) peptides produces anabolic skeletal effects that are mediated by stimulation of bone formation, thus providing a rationale for their use in glucocorticoid-induced osteoporosis. The effects of 18 months’ treatment with teriparatide [PTH (1–34), 20 μg daily] and alendronate (10 mg daily) were examined in an active-comparator randomized, double-blind, controlled trial in women and men.29 The increase in spine and hip BMD in teriparatide-treated patients was significantly greater than that in alendronate-treated patients at 6 months and 1 year, respectively. Fracture reduction was not a primary end point of this study, but radiographic evidence of new vertebral fractures was found in 10 alendronate-treated patients and 1 teriparatide-treated patient (6·1%vs 0·6%; P = 0·004). Thirty-sixth-month data from this study confirmed the superiority of teriparatide over alendronate with respect to BMD changes and a significantly reduced risk of vertebral fractures in teriparatide-treated patients when compared to those receiving alendronate.30 However, because of its higher cost and need for daily subcutaneous injection, teriparatide is generally regarded as a second-line option to bisphosphonates in the prevention and treatment of glucocorticoid-induced osteoporosis.

Calcium and vitamin.  Beneficial effects of BMD of calcium and/or vitamin D supplementation in glucocorticoid-treated patients have been demonstrated in some, but not all studies. However, calcium and vitamin D supplements have been routinely administered in most of the treatment studies and should, therefore, be used as an adjunct to treatment unless there is evidence of a sufficient dietary calcium intake and adequate vitamin D status.

Monitoring and duration of therapy.  As many individuals receiving glucocorticoids are already taking multiple medications, adherence with therapy may be compromised. It is important to explain to the patient the need for long-term bone protective therapy (in many cases) and the lack of symptomatic improvement to be expected from such treatment. Compliance with the dosing instructions for oral bisphosphonates is particularly important in glucocorticoid-treated patients because of their greater risk of upper gastrointestinal disorders. Contact with a healthcare professional 3–6 months after starting therapy provides an opportunity for the patient to discuss concerns and may improve adherence.

The role of BMD measurements in monitoring the effects of bone protective therapy in glucocorticoid-treated patients has not been well documented. However, particularly in patients receiving high doses of glucocorticoids, BMD assessment once yearly or every 2 years may be indicated, with a review of treatment in those patients who show serial losses of BMD. Similarly, although no treatment completely protects against fracture, treatment review may be indicated if more than one fracture occurs during therapy. There are very few data on the use of biochemical turnover markers in monitoring treatments for glucocorticoid-induced osteoporosis.

Because fracture risk declines quite rapidly after glucocorticoids are withdrawn, it may be appropriate to withdraw bone protective therapy when glucocorticoids are stopped. FRAX® was not designed to assess fracture risk in patients receiving bone active drugs, so risk assessment at the end of treatment is generally made on the basis of age, BMD and fracture history.

Ancillary