Basic and clinical aspects of osteoporosis in inflammatory bowel disease

Abstract

Low bone mineral density and the increased risk of fracture in gastrointestinal diseases have a multifactorial pathogenesis. Inflammatory bowel disease (IBD) has been associated with an increased risk of osteoporosis and osteopenia and epidemiologic studies have reported an increased prevalence of low bone mass in patients with IBD. Certainly, genetics play an important role, along with other factors such as systemic inflammation, malnutrition, hypogonadism, glucocorticoid therapy in IBD and other lifestyle factors. At a molecular level the proinflammatory cytokines that contribute to the intestinal immune response in IBD are known to enhance bone resorption. There are genes influencing osteoblast function and it is likely that LRP5 may be involved in the skeletal development. Also the identification of vitamin D receptors (VDRs) and some of its polymorphisms have led to consider the possible relationships between them and some autoimmune diseases and may be involved in the pathogenesis through the exertion of its immunomodulatory effects during inflammation. Trying to explain the physiopathology we have found that there is increasing evidence for the integration between systemic inflammation and bone loss likely mediated via receptor for activated nuclear factor kappa-B (RANK), RANK-ligand, and osteoprotegerin, proteins that can affect both osteoclastogenesis and T-cell activation. Although glucocorticoids can reduce mucosal and systemic inflammation, they have intrinsic qualities that negatively impact on bone mass. It is still controversial if all IBD patients should be screened, especially in patients with preexisting risk factors for bone disease. Available methods to measure BMD include single energy x-ray absorptiometry, DXA, quantitative computed tomography (QCT), radiographic absorptiometry, and ultrasound. DXA is the establish method to determine BMD, and routinely is measured in the hip and the lumbar spine. There are several treatments options that have proven their effectiveness, while new emergent therapies such as calcitonin and teriparatide among others remain to be assessed.

Key Words: Inflammatory bowel disease, Osteoporosis

INTRODUCTION

Patients with inflammatory bowel disease (IBD) are at increased risk of developing disorder in bone and mineral metabolism because of several factors, including the genetic influence, cytokine-mediated nature of the inflammatory bowel disease, the intestinal malabsorption resulting from disease activity or from extensive intestinal resection and the use of glucocorticoids to control disease activity. Apparently these disturbances may also be seen since childhood, and environmental factors such as malnutrition, immobilization, low body mass index (BMI), smoking and hypogonadism may also play a contributing role in the pathogenesis of bone loss. In IBD several studies demonstrate a negative correlation between bone mineral density (BMD) and glucocorticoid use, though there is evidence that may support the opposite. In order to answer the questions about the pathogenesis, we first have to determine the factors that are involved in this extraintestinal complication. The aim of this paper is to review the basic and molecular aspects with the clinical and therapeutic features and have an overview about the trends of the bone disease related to IBD.

EPIDEMIOLOGY

Bone mineral density is decreased in a proportion of subjects with IBD as shown by epidemiological studies. The current understanding about IBD and BMD is that the overall risk of fracture may be slightly increased in IBD patients. IBD has been associated with an increased risk of osteoporosis and osteopenia and epidemiologic studies have reported an increased prevalence of low bone mass in patients with IBD. The prevalence rates from 2% to 30% for osteoporosis (OP), from 40% to 50% for osteopenia[1] and the overall prevalence of low bone mineral density is estimated in 15%. A population-based study compared IBD patients with the general population and reported similar increases in the fracture risk between Crohn's disease (CD) and Ulcerative colitis (UC)[2] and in comparison to control patients, similar to what other population-based studies have reported[3,4]. Some series have reported that in newly diagnosed IBD patients a reduced BMD has been found and this prevalence is slightly higher in patients with CD[3,5] whereas approximately 15% of patients with CD have osteoporosis[6]. There is contrasting data from a Danish case-control study where an increase in the risk of fracture among women with CD was seen, but not men with CD or patients with UC[7,8], also another study reported that the overall fracture rate in UC was similar to that of control subjects[9]. In regard to age and gender as risk factors, elderly have the highest risk of fracturing and this increased risk is evident across all age groups[3]. Some case control studies have demonstrated that gender, age, and body weight are the major determinants of bone mineral density in patients with CD. As in healthy individuals, the combined effect of these factors account for up to 50% of the variability in bone mineral density[10]. Male sex and increasing age were considered risk factors in predicting those with osteoporosis although most series report no significant difference between the genders.

Longitudinal studies show that the BMD changes are not excessive[11,12] and there is no exclusive pattern of low BMD that involves spine of the hip. The risk of hip fracture is increased by 86% in patients with CD and by 40% in patients with UC[2]. However the hip has been reported more frequently affected than the spine[13,14]. In a study of Stockbrugger et al significant number of fractures in IBD patients as in the general osteoporotic are asymptomatic, about 14.2% of the fractures seem to be underreported[15], though it is important to mention that osteoporosis occurrence is often underestimated[15].

GENETICS

Low bone mineral density and the increased risk of fracture in all gastrointestinal diseases including IBD have a multifactorial pathogenesis. There are a number of factors that can lead to enhanced bone loss, these also include genetic factors.

LRP5

Because of the central role of osteoblasts in bone for-mation, it is easy to think there are genes influencing osteoblast function and it is likely that LRP5 may be involved in the skeletal development. The protein encoded by LRP5 is a member of the low-density lipoprotein-receptor (LDLR) gene superfamily[16] and is closely related to LRP6[17]. LRP5 is transcribed in human bone tissue as well as in numerous other tissues. There is convincing findings that deleterious (loss of function) mutations in LRP5 result in loss of function and cause bone defects such as the ones seen in pseudoglioma syndrome further supporting the critical role of this gene in skeletal integrity[18]. There is some data about the identification in normal healthy individuals of a gain of function mutation in the LDL receptor-related protein 5 (LPR5) gene resulting from a autosomal dominant high bone mass trait[19] and this gain of function mutation described in LRP5 produces increased bone mass with no adverse effect on skeletal structure, contrasting the loss of function mutation that maps to the same genomic region that contains LRP5 causes the osteoporosis pseudoglioma syndrome[20]. Polymorphisms rs491347 rs1784235 could be important to human osteoporosis phenotypes and may be considered as possible susceptibility factors for osteoporosis and fractures in humans[21]. A Japanese study found that the A1330 V polymorphism may contribute to osteoporosis susceptibility[22] and also was associated with reduced BMC and BMD values in healthy young Finnish men, providing evidence for the crucial role of LRP5 in peak bone mass acquisition[23].

VDR (Vitamin D receptor gene)

The identification of vitamin D receptors (VDRs) in peripheral blood mononuclear cells sparked the early interest in vitamin D as an immune system regulator[24]. Vitamin D deficiency has been linked to several different diseases, including the immune system-mediated OP such as IBD. The association of VDR gene BsmI polymorphism with OP has been studied by several investigators[24-28]. In addition, TaqI, FokI and ApaI polymorphisms of the VDR gene have also been described[25]. Regarding OP, most data concern to the BsmI polymorphism of the vitamin D receptor (VDR) gene.

Candidate genes

There are other candidate genes that seem involved with bone loss. Estrogen receptor alpha (ER alpha) play an important role in increasing BMD via mechanical strain and muscle mass[29]. The results of studies regarding the association between some common polymorphisms of the aromatase gene and bone mineral density and the risk of osteoporotic fractures are recognized[30]. Thus, aromatase is also an attractive osteoporosis candidate gene. The gene encoding TGFβ1 is a strong functional candidate for genetic susceptibility to osteoporosis. Several polymorphisms have been identified in TGFβ1, and previous work has suggested that allelic variants of TGFβ1 may regulate BMD and susceptibility to osteoporotic fracture[31]. During the last years, about 170 candidate genes have been published. There have been (e.g., VDR, ER-α, and COL1A1), as well as novel genes recently discovered to be important in bone and mineral metabolism. The newly studied genes include a big list CYP17 (17-hydroxylase)[32], CYP1B1 (cytochrome P450)[33], DBP (vitamin D-binding protein)[34], GH1 (growth hormone 1)[35], GnRH (gonadotropin-releasing hormone 1[35]), IGF-II (insulin-like growth factor II)[37], LEPR (leptin receptor)[38], LRP5 (low-density lipoprotein receptor-related protein 5)[39], BMP2 (bone morphogenetic protein 2)[40], CCR2 (chemokine)[41], CLCN7 (chloride channel 7)[42], COMT (catechol-O-methyltransferase)[43], CTSK (cathepsin K)[44], DRD4 (dopamine receptor D4)[45], I-TRAF (TRAF family member-associated NF-κB activator)[46], LCT (lactase)[47], MIF (macrophage migration inhibitory factor)[48], MMP-1 (matrix metalloproteinase 1)[49], among many others, but their relationship with inflammation as a possible mechanism of osteoporosis still is not clear and the interaction with IBD bone disease has not been elucidated. The mechanisms involved and the potential usefulness of those genetic data in the prevention and management of osteoporosis need further investigation, also to determine the direct relation with IBD.

PATHOPHYSIOLOGY

Inflammation has now moved to the center of the physio-pathologic mechanisms involved in the process of bone loss in IBD, there has been a considerable increase in knowledge surrounding the genetic determinants of osteoporosis. As well as genetic markers are potentially helpful in identifying high risk patients, the genetic variations of cytokines plays a key role in the regulation of the inflammatory response. Several studies are focused trying to identify genetic risk factors for rapid bone loss in IBD patients as a model of disease and inflammation-associated bone loss. Evidence accumulated in the past years support that interleukin 6 (IL-6) is a pathogenic factor in osteoporosis that results from the loss of either male or female sex steroids and have implicated IL-6 in the physiopathology of several other diseases caused by increased osteoclastic bone resorption including diseases such as Rheumatoid arthritis[50]. Genetic variations in the IL-6 and interleukin 1 receptor antagonist (IL-1ra) gene identify IBD patients at risk for increased bone loss. Allele status of the IL-1ra, IL-6, heat shock protein 70-2 and 70-hom (hsp 70-2, hsp hom) gene has been typed and correlated with clinical course of IBD and extent of bone loss[51]. These variations are independent determinants of bone loss in the setting of IBD, and have been identified as independent predictors of bone loss in the setting of postmenopausal osteoporosis, suggesting that IL-6 and IL-1ra determine the response of bone to different stressors such as the hypoestrogenic state or systemic inflammation[52,53]. Apparently, estrogen loss results in increased production of IL-6 by ex vivo bone marrow cell cultures and increased production of IL-6 follows the withdrawal of estradiol from primary culture[54,55]. It seems that IL-6 is responsible for increased bone resorption after loss of sex steroids and that gonadectomy prevents the increase in osteoclastogenesis in bone marrow and the increase in the number of osteoclasts in sections of trabecular bone[56]. The cytokines IL-1ra and IL-6 also have a central role in the paracrine stimulation of osteoclast development and regulation of the process of bone resorption[50,55]. Increasing evidence suggests that IL-6 type cytokines also promote the development of osteoblasts[50]. It has been observed that the carriage of the A2 allele of the IL-1ra gene is associated with reduced bone loss[52].

The interleukine-2 (IL-2) deficient mouse model of colitis is known to develop both osteopenia and colitis. Osteopenia was not evident in IL-2 deficient mouse cross-bred to be T-cell deficient, and osteopenia could be induced in T-cell-deficient mice by adoptive transfer of T cells from IL-2 deficient mice[57]. These data suggest that activated T cells are critical for mediating the osteopenia.

OPG-RANK-RANKL system

The receptor activator of nuclear factor κB ligand (RANKL) osteoprotegerin (OPG) system represents a potential link between inflammation and bone homeostasis and also an example of inflammation-mediated osteopenia such as IBD-associated osteopenia. The balance between RANKL and OPG (the soluble decoy receptor preventing ligation of RANKL) is of major importance to the regulation of osteoclastogenesis. The interaction of RANK on the surface of osteoclasts with its ligand RANKL induces osteoclastogenesis and conversely the interaction with the osteoblast derived soluble decoy receptor, osteoprotegerin (OPG)[58] blocks RANK-RANKL interaction inhibiting osteoclasts formation. Whether compounds stimulate RANK ligand or OPG will affect whether they induce or inhibit osteoclastogenesis. Pro-inflammatory cytokines induce RANKL and promote bone resorption with consecutive bone loss. Activated T cells can directly trigger osteoclastogenesis through RANKL leading to bone loss while OPG can block those effects[59-61]. Increased OPG levels may represent a continuing homeostatic response, attempting to reverse established osteopenia and RANKL driven osteoclastogenesis, thus maintaining normal bone mass. Inflammation seems to play an important role in the regulation of the OPG-RANK-RANKL system. To correlate it with chronic inflammatory states comparable to IBD, there have been some reports that show a direct correlation between serum OPG and erythrocyte sedimentation rate and a score of disease activity in patients with rheumatoid arthritis[62]. Soluble RANKL as well as OPG levels are elevated in rheumatoid arthritis, while high OPG and decreased RANKL levels have been reported in primary biliary cirrhosis[63,64]. Some of the osteoclastogenic factors released from the IBD mucosa (for example IL-1, IL-6 and TNFα) are thought to function indirectly via specific receptors on stromal osteoblastic cells to enhance RANKL expression[60,65,66]. Data suggests that OPG may be a protective host response that partially offsets the adverse skeletal effect created by the inflammation state. Moshen et al[67] described the alterations in the RANKL/OPG system in IBD and its relationship to decreased BMD. It has been demonstrated increased plasma levels of OPG as well as increased release from the inflamed colon in IBD, suggesting the macrophages and dendritic cells as colonic source of OPG in IBD. Apparently, no correlation was evident between corticosteroid and serum OPG[63] contrasting partially with other findings.

Corticosteroids

The controversial participation of glucocorticoid (GC) therapy in the pathogenesis of bone loss in IBD still has gaps to be fulfilled. It seems that there is an important relationship between dosage, duration and pattern of GC therapy and these factors are related to the incidence of pathological fractures[68]. Some studies indicate that fractures are present in 30%-50% of patients on GC therapy for chronic diseases[69] and several studies have demonstrated that dosage is associated with BMD[51,70-73]. On the other hand, several studies have reported the opposite[8,13].

The epidemiological data on fracture risk and bone loss in GC therapy do not distinguish the effects of drug and the effects of the underlying disease. It is known, for example, in rheumatoid arthritis, the risk of fracture is increased even in the absence of GC exposure, also it has been observed that osteoporosis is rapidly developed in recently diagnosed Crohn’s disease without any effect of corticosteroids in the follow up. One study showed that the prevalence of osteoporosis in pediatric patients with IBD is approximately the same as in adult patients, showing that osteoporosis was already present before steroid treatment[74]. Contrasting data from other studies show that the extent of bone loss was no correlated to clinical severity of disease or application of corticosteroids[75-77]. The participation of GC in the pathophisiology of bone loss is complex. GCs influence the production and action of hormones that regulate bone and calcium metabolism and also have direct effects of GCs on bone. GCs increase the expression of receptor activator of nuclear factor κB ligand (RANK-L) and decrease the expression of its soluble decoy receptor osteoprotegerin (OPG) in stromal and osteoblastic cells[78] and also enhance the expression of macrophage colony-stimulating factor (M-CSF), which in the presence of RANK-L induces osteoclastogenesis[78-80]. GCs have direct effects on osteoclasts also by suppressing the expression of an autocrine cytokine, such as interferonI, that normally exerts inhibitory effects on osteoclastogenesis[80]. Also they inhibit the function of mature osteoblasts and suppress the synthesis of insulin-like growth factor-I, an agent that enhances bone formation[78,79].

The wingless-type (Wnt) signaling has emerged as a novel, key pathway for promoting osteoblastogenesis. The Wnt signal transduction comprises three intracellular pathways: the canonical pathway, the Wnt/planar-cell-polarity (PCP) pathway, and the Wnt/Ca²⁺ pathway[81,82]. Wnt signals are extracellularly regulated by several secreted antagonists including secreted frizzled-related protein (sFRP), Cerberus, Wnt inhibitory factor-1 (WIF-1), and dickkopf (Dkk)[83]. Some studies strongly suggest that the canonical pathway plays a central role in promoting bone formation[84-86]. Some groups have reported that glucocorticoid enhances the expression of dickkopf-1 (Dkk-1) in cultured human osteoblasts[87] by suppressesing the canonical Wnt signal[88].

DIAGNOSIS

Diagnosis of osteoporosis in IBD patients

Due to the low absolute risk of fracture remains contro-versial if all IBD patients should be screened, but it is suggested for avoiding the complications of osteoporosis, especially in patients with a preexisting bone disease, older than 65, and with risk factors for low bone mass as long-term steroid therapy (prednisone 5 mg daily for 6 mo or more)[88-91].

Both, the American College of Gastroenterology (ACG) and American Gastroenterological Association (AGA) issued position papers to offer guidance to the practicing clinician in the diagnosis and management of bone loss in IBD. These position papers recommended the selective screening of IBD patients with dual energy x-ray absorptiometry (DXA) scanning, and the criteria for DXA screening included: postmenopausal state , ongoing corticosteroid treatment, cumulative prior use of corticosteroids exceeding 3 mo, history of low trauma fractures, and age over 60. These criteria led to the detection of osteopenia or osteoporosis and initiation of specific therapies in the majority of patients[92].

Available methods to measure BMD include single energy x-ray absorptiometry, DXA, quantitative computed tomography (QCT), radiographic absorptiometry, and ultrasound. DXA is the establish method to determine BMD, and routinely is measured in the hip and the lumbar spine[93].

The T score was proposed by the World Health Organization (WHO) as the strongest determinant of fracture risk. T score is defined as the number of standard deviations (SD) by which a given BMD measurement exceeds or falls below the normal mean BMD of healthy 30-year-old individuals (peak bone mass). A BMD that is up to 1 SD below the peak bone mass is considered normal; between 1 to 2.49 SD below peak BMD is considered as osteopenic and to have mild to moderate bone deficiency; and ≥ 2.5 SD below the peak BMD are labeled osteoporotic and with marked bone deficiency. Individuals who have a fracture as a result of bone fragility are considered to have severe osteoporosis[93]. The z score is useful too, and is defined as the number of SDs by which a given BMD measurements exceeds or falls below the mean BMD of healthy individuals of the same age group. For the International Society for Clinical Densitometry (ISCD), z scores are preferred, and the WHO classification should not be applied in women before menopause and in men younger than 50[94].

TREATMENT OPTIONS

Calcium and vitamin D

It is known that calcium and vitamin D are essential in the metabolism of bone and so multiple trials have studied their benefit as treatment of osteoporosis. The use of calcium or/and vitamin D or its analogues have shown, in 2 meta-analysis, 1 Cochrane Review, and in a large placebo-controlled study, to have a small benefit in BMD and a controversial age-dependant trend, and not totally clear reduction of bone fractures, specially of the spine, in postmenopausal women[95-98]. In a randomized, placebo-controlled trial in glucocorticoid-using patients with IBD, the intake of vitamin D 250 IU and calcium 1000 mg/d had no significant benefit in bone density at 1 year of follow-up[99]. So, as described in a recent consensus report, the supplementation with calcium and vitamin D is accepted as a cost-effective medication, and essential but insufficient, in the prevention and treatment of osteoporosis. The dosage that showed best is calcium 1200 mg/d and vitamin D 800 IU/d. The maximum benefit of calcium and vitamin D will generally be derived from combination therapy with an antiresorptive agent[100].

Bisphosphonates

The group of this antiresorptive analogue of pyrophosphate includes etidronate, pamidronate, tiludronate, alendronate, risedronate, and ibandonate.

Both, alendronate and risedronate, have shown to be effective in increasing BMD and reducing fractures in spine, hip and wrist for the treatment of osteoporosis in postmenopausal women. In a systematic review, meta-analysis and double blind and randomized study, they reduce vertebral fractures by 30% to 50%, with superiority for 70 mg once-weekly alendronate than daily 5 mg or once-weekly 35 mg of risedronate, and with similar tolerability profiles, at 1 or 2 years[101-105].

For the prevention and treatment of glucocorticoid-induced osteoporosis, in a randomized, double-blind, placebo-controlled, multicenter study, in patients receiving a minimum of 7.5 mg prednisone or its equivalent for diverse pathologies, all receiving 800-1000 mg elemental calcium and 250-500 IU of vitamin D, alendronate at a dosage of 5 or 10 mg/d significantly increased bone density compared to placebo at 1 year and reduced the incidence of bone fractures too, at 2 years[106,107].

In patients with moderate to high doses of corticoid therapy, a significant increase of BMD and a reduction of 70% in vertebral fracture risk was observed with risedronate 5 mg/d compared with the placebo group (P = 0.01). Risedronate was efficacious, irrespective of underlying disease and duration of corticosteroid therapy, and had a favorable safety profile, with a similar incidence of upper gastrointestinal adverse events to placebo[108,109].

Etidronate have shown to be superior to placebo for increasing BMD in lumbar spine and femoral neck, and reducing incidence of vertebral fractures with no effect in non-vertebral fractures in postmenopausal women[110].

A meta-analysis reported that intermittent cyclical etidronate (400 mg/d for 14 d, followed by 500 mg calcium daily for 76 d) in corticoid treated patients was effective in preventing bone loss, increasing bone mass but with no statistical significance on reduction of fractures[111].

Other bisphosphonate approved for the treatment of osteoporosis in postmenopausal women is the ibandonate in oral dosage of 2.5 mg/d, or intravenous dosage of

2 mg every 2 mo, or 3 mg every 3 mo, had shown to be better than placebo, increasing BMD and reducing bone fractures, with superiority of intravenous regimens[112].

For corticoid-induced osteoporosis, in an open-label, single-center, parallel-group, controlled study, participants received 500 mg/d calcium plus either 3-montly intravenous injections of 2 mg ibandronate or oral 1 mg/d alfacalcidol for 3 years, showing that the increase in BMD was much greater and the fractures were lower in the ibandronate than those in alfacalcidol group[113].

For the treatment of osteoporosis in IBD, bisphos-phonates have been evaluated in few studies. In a 12-mo double-blind, randomized, placebo-controlled study of 10-mg daily dose of alendronate, that include 32 patients with CD in remission and without glucocorticoid treatment the BMD of the lumbar spine increased 4.6% ± 1.2% versus a decrease of 0.9% ± 1.0% in the placebo group (P < 0.01).

BMD of the hip increased 3.3% ± 1.5% vs an increase of 0.7% ± 1.1% in the placebo group (P < 0.08)[114].

In 31 patients with CD and 30 with UC, in a double-blind placebo-controlled study, all taking 600 mg daily of calcium, after 1 year in the risedronate group the BMD of the spine and hip significantly increase in 2% and 1.9%, respectively[115]. After one year of monthly infusions of 30 mg iv pamidronate plus 500 mg calcium with 400 IU vitamin D in patients with CD, the BMD increased 2.6% (95% CI: 1.4-3.0) at the spine and 1.6% (95% CI: 0.6-2.5) at the hip versus 1.6% (95% CI: 0.1-3.2) at the spine and 0.9% (95% CI: 0.4-2.1) at the hip in the group with vitamin D and calcium supplements[116]. Stokker PC et al [117] reported a significant improve in T scores of lumbar spine and hip in 49 patients with IBD that received 30 mg iv pamidronate every 3 mo, plus 1000 mg of calcium and 400 IU of vitamin D daily.

Estrogens

Estrogens alone or with progestin stop progression of bone loss in postmenopausal women, increasing the BMD and reducing the incidence of spine and hip fractures by 34%[118]. Good response in preventing bone loss in patients under glucocorticoid treatment has been observed but the effect on prevention of bone fractures remains unclear, estrogens are not recommended for this purpose[119,120].

Raloxifene, a selective estrogen receptor modulator was approved for the prevention and treatment of postmenopausal spinal osteoporosis. In a meta-analysis of 7 clinical studies, raloxifene reduced the risk of vertebral fractures by 40% with a dose of 60 mg/d[121]. No studies with raloxifene have done yet in IBD patients.

EMERGENT THERAPIES

Calcitonin

Calcitonin intranasal spray, at doses of 200 IU/d plus 1000 mg calcium and 400 IU vitamin D, has been reported to reduce the risk of spine fractures by 33% in a 5-year follow-up time in postmenopausal women[122].

The efficacy of calcitonin for fracture prevention in steroid-induced osteoporosis remains to be estab-lished[123,124]. No studies have done for IBD-associated osteoporosis.

Teriparatide

The genetically engineered fragment of human parathyroid hormone, Teriparatide, stimulates new bone formation, leading to increased BMD. Teriparatide, at 20 and 40 micrograms daily subcutaneous injection, reduced the risk of vertebral and non-vertebral fractures in postmenopausal women[125]. It’s also approved for FDA to increase bone mass in men with primary or hypogonadal osteoporosis[126].

The efficacy of teriparatide in preventing of treating glucocorticoid-induced or IBD-associated osteoporosis remains to be assessed. Hodsman AB[127] suggests that should be considered as treatment for patients with established glucocorticiod-induced osteoporosis who require long-term steroid treatment.