PerspectivePerspectives on glucocorticoid-induced osteoporosis
Introduction
Glucocorticoid-induced osteoporosis (GIO) is the most common cause of secondary osteoporosis and is the result of the profound effects of glucocorticoids on the skeleton. This article is a perspective on new discoveries at the cellular and clinical level as they relate to GIO. Some of the concepts outlined are derived from selected presentations made at the 3rd International Congress on GIO held in Turin, Italy, in March 2003.
The systemic effects of glucocorticoids on skeletal metabolism are relevant to their ultimate effects on the skeleton. However, recently, our attention has focused on the direct actions of these steroids on skeletal cells. Bone is continuously remodeled by basic multicellular units composed of teams of juxtaposed osteoclasts and osteoblasts. Alterations in the number of cells or in their function by systemic or local agents can lead to an imbalance in skeletal homeostasis and eventually to bone loss. Glucocorticoids have direct effects in bone cells leading to changes in cell number and function that in the aggregate account for bone loss.
Patients exposed to glucocorticoid therapy exhibit a phase of increased bone resorption. Prior emphasis to explain this event had been placed on effects of glucocorticoids on other organs involved in calcium metabolism [1]. These effects include inhibition of calcium absorption in the gastrointestinal tract and induction of renal calcium loss. Although these effects could explain, in an indirect manner, increased bone resorption, recent data suggest a more direct effect of glucocorticoids on bone cells. Glucocorticoids increase the expression of receptor activator of NF-kappa B ligand (RANK-L) and decrease the expression of its soluble decoy receptor, osteoprotegerin, in stromal and osteoblastic cells [2]. Glucocorticoids also enhance the expression of colony-stimulating factor (CSF)-1, which in the presence of RANK-L induces osteoclastogenesis [3]. These actions likely explain the increased bone resorption that follows skeletal exposure to glucocorticoids. These are early events. Eventually, a more chronic state of decreased bone remodeling develops, which is secondary to a loss of cells signaling to osteoclasts or to their progenitors, and to the apoptosis of mature osteoclasts [4]. However, under selected experimental conditions, corticosteroids were found to extend the life of the osteoclast, and to oppose the effect of bisphosphonates on osteoclast apoptosis [5]. This indicates that, depending on the presence of other signals present in the bone environment, glucocorticoids can favor or oppose bone resorption.
Among the mechanisms by which glucocorticoids induce bone resorption, a hyperparathyroid state has been considered to be of importance. Earlier literature showed increases in serum levels of parathyroid hormone (PTH) when patients exposed to chronic glucocorticoids were studied [6], [7]. Other postulates have included enhanced sensitivity to PTH due to changes in the number and affinity of PTH for its receptors [8]. If these mechanisms, leading to a hyperparathyroid state in GIO, are present, one should expect consistent elevations of serum PTH and a pattern of bone loss that mirrors that seen in hyperparathyroidism with histomorphometric evidence for excessive PTH action in bone [9].
Acute or chronic use of glucocorticoids is not regularly associated with elevated endogenous levels of PTH [10], [11]. Furthermore, PTH secretory dynamics are altered by glucocorticoids with a reduction in the tonic component and an exaggeration in the pulsatile component of PTH secretion that override normal secretory dynamics [12]. These pulsatility studies call attention to the need to consider not only the amount of PTH secreted, but also to its pattern of secretion in the presence of glucocorticoids. Bone densitometric studies also introduce a cautionary note with respect to the involvement of PTH in GIO. The typical bone densitometric pattern in primary hyperparathyroidism, the prototypic disorder of PTH excess, is preferential bone loss in the cortical skeleton with relative preservation of bone mass in the cancellous skeleton [13], [14]. This pattern is seen commonly in patients with asymptomatic primary hyperparathyroidism, illustrating well the predilection of the cortical skeleton to be affected by the catabolic actions of PTH. In contrast, in GIO, an opposite pattern is found since glucocorticoids are associated with preferential loss of cancellous bone with a major loss of lumbar spine bone density before the peripheral skeleton is involved [15], [16]. As expected, reductions in cancellous bone density in GIO are associated with an increased risk of fractures of the vertebral spine. This pattern of bone loss contrasts with that seen in primary hyperparathyroidism, where vertebral compression fractures are uncommon. In addition, histomorphometric analysis of bone biopsy specimens in GIO exhibits slower bone turnover with a disproportionate reduction of bone formation over bone resorption [17]. In primary hyperparathyroidism, histomorphometric patterns are distinctly different revealing enhanced bone turnover with preservation of trabecular bone [18]. Cellular processes associated with GIO are different from those seen in primary hyperparathyroidism, and glucocorticoids suppress osteoblast number and activity, whereas PTH does not [19], [20]. Altogether, the evidence has clearly shifted away from the notion of secondary hyperparathyroidism in GIO. On the contrary, PTH would appear to have no major role in the pathogenesis of GIO.
Although glucocorticoids regulate bone resorption, the general consensus, based on bone histomorphometric studies, favors decreased bone formation as the most significant event leading to chronic glucocorticoid-induced bone loss [17]. A significant consequence of the skeletal exposure to glucocorticoids is a decrease in the number of cells of the osteoblastic lineage [1], [19]. Cell genesis and death are the ultimate determinants of the pool of osteoblasts available to form bone. Therefore, it is not surprising that glucocorticoids have an impact on osteoblastogenesis and osteoblastic apoptosis [19], [21]. Glucocorticoids induce the apoptosis of osteoblasts and osteocytes, and this contributes to a decreased number of mature osteoblasts [22]. Recent attention has focused on the effects of glucocorticoids on osteoblastogenesis and the mechanisms involved in a shift of cellular differentiation away from osteoblasts and toward adipocytes [21]. Although some investigators have reported that glucocorticoids induce osteoblastic cell differentiation, this is inconsistent with the loss of cells of the osteoblastic lineage and of osteoblastic function observed after glucocorticoid exposure [23]. In fact, glucocorticoids impair the differentiation of stromal cells toward cells of the osteoblastic lineage and prevent the terminal differentiation of quasi-mature osteoblastic cells, resulting in a decrease in the number of mature osteoblasts [19], [21]. Glucocorticoids shift the differentiation of stromal cells toward the adipocytic lineage. This shift involves the regulation of nuclear factors of the CCAAT/enhancer binding protein (C/EBP) family and of peroxisome proliferator-activated receptor γ2 (PPARγ2) [24], [25]. Of the six C/EBPs identified, C/EBP α, β, and δ play essential roles in adipogenesis. However, the shift of stromal cells away from the osteoblast and toward adipocyte differentiation caused by glucocorticoids appears to involve additional cellular signals.
Recently, it was demonstrated that cortisol induces Notch1 mRNA levels in osteoblasts, and Notch1 plays a role in adipogenesis [26], [27]. Notch consists of a family of four transmembrane receptors activated by their ligands Delta and Jagged [28], [29]. Notch1 and 2 and their ligands, Delta 1 and Jagged 1, are expressed by osteoblasts, whereas Notch3 and 4 are not [26], [30]. Cortisol increases Notch1 and 2 transcripts, but does not modify the expression of its ligands. Overexpression of Notch1 in stromal and osteoblastic cells mimics some of the effects of glucocorticoids, impairing osteoblastic maturation and favoring adipogenesis [27]. It is of interest that Notch1 and Wnt have opposite effects on cell differentiation and adipogenesis, and Notch1 overexpression in stromal cells opposes Wnt/β-catenin signaling [27], [31], [32], [33]. If Notch1 induction were to play an intermediate role in the action of glucocorticoids in osteoblasts, then it is conceivable that an ultimate event would be inhibition of Wnt signaling by these steroids.
Selected actions of glucocorticoids are secondary to the regulation of the growth hormone/insulin-like growth factor (IGF) axis [34]. IGF I has stimulatory effects on bone formation, opposite to those of glucocorticoids, and its skeletal levels are decreased by cortisol, which regulates the binding of C/EBPs to a recognition site adjacent to the third start site of transcription [35]. The involvement of C/EBPs in the regulation of IGF I expression and in the effect of glucocorticoids on adipogenesis reveals that the effects of these steroids converge on specific cellular signals to regulate cell fate and function.
Individual susceptibility to glucocorticoids varies considerably. Studies in vitro indicate that at physiological levels, the effects of glucocorticoids upon bone are diverse and are regulated at the pre-receptor level in an autocrine fashion [36], [37]. Pre-receptor steroid regulation is critical to steroid action and it is not unique to glucocorticoids. Androgens and estrogens are also regulated by local enzymes, including 5α reductase, hydroxysteroid dehydrogenase, and aromatase [37]. 11β-hydroxysteroid dehydrogenases (11β-HSDs) are two isoenzymes that catalyze the interconversion of hormonally active cortisol and inactive cortisone, and are therefore critical in the regulation of glucocorticoid activity. 11β-HSD1, a low affinity NADP(H)-dependent enzyme, is bidirectional (dehydrogenase/reductase), but it displays primarily reductase activity and converts cortisone to cortisol [36]. 11β-HSD1 acts as a pivotal determinant of steroid responses in bone by amplifying glucocorticoid signaling in osteoblasts. At a clinical level, recent analysis of age-specific variations in osteoblastic 11β-HSD1 activity suggests that this mechanism is a contributing factor in age-related and glucocorticoid-induced bone loss [38]. Bone histomorphometric analysis of biopsies from patients receiving glucocorticoids reveals decreased bone remodeling, a state more often associated with senile and not postmenopausal osteoporosis [17], [39]. Recently, an age-related shift toward type I activity of osteoblast 11β-hydroxysteroid dehydrogenase has been advocated as an additional intracellular mechanism leading to enhanced cortisol availability, hence enhanced glucocorticoid effects in the skeleton of elderly subjects [37], [38].
11β-HSD1 is widely expressed in glucocorticoid target tissues, including bone. Since 11β-HSD1 activates inactive glucocorticoids, it can facilitate glucocorticoid action in target tissues. The activity and synthesis of 11β-HSD1 in osteoblasts is glucocorticoid dependent [38]. Consequently, it may serve as a positive local mechanism to amplify the effect of glucocorticoids in this target tissue. In addition, proinflammatory cytokines, often present in excess because of the underlying disease being treated with glucocorticoids, can modulate 11β-HSD1 and amplify the effect of these steroids in bone [40]. The mechanism determining the induction of 11β-HSD1 by glucocorticoids in osteoblasts is not known, but could involve the C/EBP family of transcription factors since transient transfections of C/EBP α result in transactivation of the 11β-HSD1 promoter, which contains numerous C/EBP binding sites [41]. C/EBP β acts as a negative regulator of 11β-HSD1 transcription, possibly through the formation of heterodimers with C/EBP α [40].
There is a growing body of evidence from epidemiological studies that glucocorticoid use is associated with accelerated bone loss and osteoporotic fractures [42]. Despite this, glucocorticoids are used widely in the practice of medicine. A recent study, using the General Practice Research Database, a detailed pharmacological recording system that covers some 7 million individuals in England and Wales, identified 1.6 million oral glucocorticoid prescriptions over a 10-year period [43]. At any one time, the prevalence of oral glucocorticoid use was 0.9% of the total adult population. The use of oral glucocorticoids varies substantially with age, being higher in the elderly population, a group already at risk for osteoporosis. It is important to note that database studies have demonstrated that even patients receiving what have been considered physiological doses of glucocorticoids (prednisolone equivalents 2.5–7.5 mg/daily) are at an increased risk of vertebral fractures [44]. Interestingly, this is the most commonly used dose of oral corticosteroids. Admittedly, these epidemiologic studies may be confounded by the fact that those on glucocorticoids, even at low doses, may have disorders for which osteoporosis is a direct consequence.
Although glucocorticoids are the most common cause of secondary osteoporosis, diagnostic thresholds in GIO, using bone mineral density (BMD), have not been established. It is important to recognize that diagnostic guidelines for postmenopausal osteoporosis established by the World Health Organization do not apply to GIO. Patients exposed to glucocorticoids are at an increased risk of fractures. At similar BMD levels, glucocorticoid users appear to have higher risk of fractures than nonusers, a reason why the guidelines of the Royal College of Physicians of London have proposed to consider a T score of −1.5 or lower to be indicative of a need of therapeutic intervention [42], [45]. The higher risk of fractures at comparable BMDs points to issues of bone quality that may be affected by glucocorticoids, but not measured by the bone density test. Although our awareness regarding GIO has increased and attitudes toward the use of glucocorticoids in clinical practice have changed in recent years, only a fraction of patients on continuous glucocorticoid therapy receives any treatment to prevent bone loss. A study on practice attitudes in an academic setting in the United States revealed that only one-quarter of patients with rheumatic diseases receiving continuous glucocorticoid therapy underwent diagnostic testing with BMD and received supplemental calcium and vitamin D [46].
Patients receiving glucocorticoids have an underlying disease, which frequently by itself carries a risk of osteoporosis. These underlying disorders include rheumatologic diseases, as well as chronic pulmonary disorders, inflammatory bowel disease (IBD), and transplant patients [47]. Patients with chronic obstructive pulmonary disease (COPD) are at increased risk of osteoporosis due to a variety of factors associated with the disease, such as poor health, poor nutrition, and tobacco consumption, among others [48]. Inflammatory bowel disease (IBD), a term that encompasses ulcerative colitis and Crohn’s disease, is associated with osteopenia in 30–50% of patients [49]. The pathogenesis of bone loss associated with inflammatory bowel disease is multifactorial. Glucocorticoid therapy, hypogonadism, vitamin D deficiency or insufficiency, malnutrition, and low body weight are all important contributory factors. In addition, cytokines released by the inflamed intestinal mucosa may play a role [50]. Histomorphometric studies in IBD are sparse, but generally indicate that reduced bone formation is the predominant event. However, it is likely that increased bone turnover also contributes to bone loss in some patients, as indicated by the measurement of biochemical markers in these patients. There have been few studies of therapeutic interventions to prevent bone loss in patients with IBD, and these patients have been poorly represented in trials of GIO prevention and treatment [51]. The identification of effective treatment strategies in these patients is thus an important priority for future research. Transplant recipients are at increased risk of osteoporosis due to the use of glucocorticoids, immune suppressive therapy, and the underlying disease [52].
Patients discontinuing systemic therapy exhibit a trend toward recovery of BMD [44]. Whenever possible, the use of systemic glucocorticoids should be minimized. Similarly, the successful surgical treatment of patients with Cushing's syndrome is associated with some restoration of bone mass. A common question in clinical practice is whether patients with respiratory disorders are at risk following the use of inhaled steroids. For the most part, prospective and retrospective studies carried out in patients with asthma or COPD indicate that inhaled steroids carry a modest risk of osteoporosis and osteoporotic fractures [48]. The cumulative dose and length of exposure seem to play a significant role in the modest loss of bone reported [53], [54]. The assessment of their impact on patients with COPD is more difficult due to multiple confounding factors affecting patients with the disease [48].
Patients exposed to glucocorticoids should receive supplemental calcium and vitamin D. In addition, current evidence and clinical guidelines favor pharmacological intervention for the prevention and treatment of GIO [55], [56]. Prevention studies have examined therapeutic effects on individuals exposed to glucocorticoid therapy for <3 months. There is evidence that etidronate, alendronate, risedronate, clodronate, and intravenous pamidronate all prevent the decline in BMD that follows the initial treatment with glucocorticoids [57], [58], [59], [60]. Treatment trials have examined the effect of the therapeutic intervention on individuals exposed to glucocorticoids for periods of >6 months. Lumbar spine BMD decreased or remained stable in the placebo arms and was increased by alendronate or risedronate treatment [45], [58]. Although there was a trend to a reduction in vertebral fractures by alendronate in the GIO trial, there were few patients with new vertebral fractures in the study, so that the effect was not significant after 1 year. When the risedronate prevention and treatment studies were pooled, a significant reduction of 70% in vertebral fractures was seen in the risedronate arm after 1 year [45], [59].
Human PTH 1–34, or teriparatide, has anabolic actions in bone and was shown to increase BMD of the spine and hip in postmenopausal women on glucocorticoids. In this study by Lane et al. [61], the patients had been and continued to receive estrogen hormonal therapy (HT). The effect on BMD was sustained for 1 year following the discontinuation of PTH and the continuation of HT.
The premenopausal woman treated with glucocorticoids is of particular concern. Patients who maintain normal ovarian hormonal function seem to have a degree of protection from the deleterious effects of corticosteroids, although they may develop osteoporosis [45], [58]. A concern with the use of bisphosphonates in premenopausal women is the fact that they cross the placenta and could inhibit fetal bone remodeling and have teratogenic effects [62]. Their prolonged half-life in bone is also of concern in premenopausal women. It is therefore only under selected circumstances, where the benefits clearly outweigh the risks, that bisphosphonate therapy would be used in women of childbearing age. An alternative could be the use of PTH or calcitonin, although the effectiveness of calcitonin in GIO is not clear.
Guidelines published by the American College of Rheumatology (ACR) and the Royal College of Physicians of London advocate the following measures for prevention and treatment of GIO: general health awareness, the administration of calcium and vitamin D, reduction of the dose of corticosteroids to a minimum, and therapeutic intervention with bisphosphonates or alternate agents [42], [55]. In subjects who are to receive glucocorticoids for at least 3 months, bone densitometry should be considered. The ACR considers a T score of −1 or lower and the Royal College of Physicians of London a T score of −1.5 or lower to be indicative of a need for therapeutic intervention, although individual decisions should determine the management of each patient exposed to corticosteroids.
Future research in basic mechanisms of glucocorticoid action in bone and on the use of selective glucocorticoid receptor modulators or of nitrosylated derivatives of prednisolone may result in new approaches to the management of GIO [63], [64]. Another important area of future research will be the mechanisms of corticosteroid-induced myopathy, a serious complication that may lead to falls and fractures. Recent work demonstrating upregulation of myostatin, a negative regulator of muscle mass, by dexamethasone offers new information on possible mechanisms involved [65].
Section snippets
Acknowledgements
This work was supported by grants DK45227 (Ernesto Canalis) and DK32333 (John P. Bilezikian) from the National Institutes of Diabetes, Digestive and Kidney Diseases, and grants from GISGO and Centro di Ricerca sull'Osteoporosi-University of Brescia/EULO (A. Giustina). The authors thank Ms. Nancy Wallach for secretarial assistance.
References (65)
- et al.
Cortisol inhibits the differentiation and apoptosis of osteoblasts in culture
Bone
(2001) - et al.
Effects of cortisol and bone morphogenetic protein-2 on stromal cell differentiation: correlation with CCAAT-enhancer binding protein expression
Bone
(2002) Review: the ins and outs of notch signaling
Mol. Cell. Neurosci.
(1997)- et al.
Notch signaling: from the outside in
Dev. Biol.
(2000) - et al.
Localization of ADAM10 and Notch receptors in bone
Bone
(1999) - et al.
Notch-1 controls the expression of fatty acid-activated transcription factors and is required for adipogenesis
J. Biol. Chem.
(1997) - et al.
Evidence for two distinct syndromes of involutional osteoporosis
Am. J. Med.
(1983) - et al.
C/EBP regulates hepatic transcription of 11beta-hydroxysteroid dehydrogenase type1. A novel mechanism for cross-talk between the C/EBP and glucocorticoid signaling pathways
J. Biol. Chem.
(2000) - et al.
Inhaled corticosteroid use and bone-mineral density in patients with asthma
Lancet
(2000) - et al.
Management of glucocorticoid induced osteoporosis in premenopausal women with arthritis
Autoimmun. Rev.
(2003)
DNA binding of the glucocorticoid receptor is not essential for survival
Cell
Mechanisms of glucocorticoid action in bone: implications to glucocorticoid-induced osteoporosis
J. Clin. Endocrinol. Metab.
Stimulation of osteoprotegerin ligand and inhibition of osteoprotegerin production by glucocorticoids in human osteoblastic lineage cells: potential paracrine mechanisms of glucocorticoid-induced osteoporosis
Endocrinology
Dexamethasone promotes expression of membrane-bound macrophage colony-stimulating factor in murine osteoblast-like cells
Endocrinology
Glucocorticoids inhibit bone resorption by isolated rat osteoclasts by enhancing apoptosis
J. Endocrinol.
Promotion of osteoclast survival and antagonism of bisphosphonate-induced osteoclast apoptosis by glucocorticoids
J. Clin. Invest.
Altered mineral metabolism in glucocorticoid-induced osteopenia
J. Clin. Invest.
The effect of long- and short-term corticosteroids on plasma calcitonin and parathyroid hormone levels
Calcif. Tissue Int.
The role of parathyroid hormone in the pathogenesis of glucocorticoid-induced osteoporosis: a re-examination of the evidence
J. Clin. Endocrinol. Metab.
Intact parathyroid hormone levels are not elevated in glucocorticoid-treated subjects
J. Bone Miner. Res.
Parathyroid hormone and 25-hydroxyvitamin D levels in glucocorticoid-treated patients
Clin. Endocrinol.
Chronic glucocorticoid treatment alters the spontaneous pulsatile parathyroid hormone (PTH) secretory pattern in humans
Skeletal disease in primary hyperparathyroidism
J. Bone Miner. Res.
Differential effects of glucocorticoids on cortical appendicular and cortical vertebral bone mineral content
Calcif. Tissue Int.
Lateral spine densitometry is a more sensitive indicator of glucocorticoid-induced bone loss
J. Bone Miner. Res.
Comparison of trabecular bone microarchitecture and remodeling in glucocorticoid-induced and postmenopausal osteoporosis
J. Bone Miner. Res.
The histomorphometry of bone in primary hyperparathyroidism; preservation of cancellous bone structure
J. Clin. Endocrinol. Metab.
Insulin-like growth factor I mediates selected anabolic effects of parathyroid hormone in bone cultures
J. Clin. Invest.
Inhibition of osteoblastogenesis and promotion of apoptosis of osteoblasts and osteocytes by glucocorticoids
J. Clin. Invest.
Physiological concentrations of glucocorticoids stimulate formation of bone nodules from isolated rat calvaria cells in vitro
Endocrinology
Cited by (201)
An Overlooked Medication-Induced Celiac Flare Complicating Treatment of Osteoporosis
2024, AACE Clinical Case ReportsBone mineral density and skeletal outcomes in congenital adrenal hyperplasia
2023, Genetic Steroid Disorders: Second EditionGlucocorticoid-induced osteoporosis and Cushing’s syndrome
2020, Marcus and Feldman’s OsteoporosisEffects on bone health of glucocorticoid replacement therapy in primary and secondary adrenal insufficiency: A review
2018, Current Opinion in Endocrine and Metabolic Research