MANAGEMENT OF ENDOCRINE DISEASE: Secondary osteoporosis: pathophysiology and management

in European Journal of Endocrinology
Correspondence should be addressed to F Mirza; Email:

Osteoporosis is a skeletal disorder characterized by decreased mass and compromised bone strength predisposing to an increased risk of fractures. Although idiopathic osteoporosis is the most common form of osteoporosis, secondary factors may contribute to the bone loss and increased fracture risk in patients presenting with fragility fractures or osteoporosis. Several medical conditions and medications significantly increase the risk for bone loss and skeletal fragility. This review focuses on some of the common causes of osteoporosis, addressing the underlying mechanisms, diagnostic approach and treatment of low bone mass in the presence of these conditions.


Osteoporosis is a skeletal disorder characterized by decreased mass and compromised bone strength predisposing to an increased risk of fractures. Although idiopathic osteoporosis is the most common form of osteoporosis, secondary factors may contribute to the bone loss and increased fracture risk in patients presenting with fragility fractures or osteoporosis. Several medical conditions and medications significantly increase the risk for bone loss and skeletal fragility. This review focuses on some of the common causes of osteoporosis, addressing the underlying mechanisms, diagnostic approach and treatment of low bone mass in the presence of these conditions.

Invited Author's profile

F Mirza is Assistant Professor of Medicine, University of Connecticut School of Medicine. She specializes in the management of patients with metabolic bone disease, osteoporosis, diabetes, thyroid and pituitary disorders, and women's health issues.

Invited Author's profile

E Canalis is Professor, Departments of Orthopaedic Surgery and Medicine, UConn Health. The Canalis laboratory discovered the existence of skeletal growth factors and his interests center around growth factors and their antagonists, anabolic agents and the management of osteoporosis.


Osteoporosis is a skeletal disorder characterized by low bone mass and microarchitectural deterioration of the skeleton leading to bone fragility and a predisposition to fractures. Osteoporosis is a major cause of morbidity and mortality, particularly in post-menopausal women and older men. While the pathogenesis of the bone loss and skeletal fragility is not well understood, estrogen deficiency plays a role in both sexes. It is classified as primary osteoporosis when it occurs in post-menopausal women and in men in the absence of an underlying disease, and it is age-related. Secondary osteoporosis is defined as low bone mass with microarchitectural alterations in bone leading to fragility fractures in the presence of an underlying disease or medication (1).

Secondary osteoporosis can be present in pre- and post-menopausal women and in men. Up to 30% of post-menopausal women and 50–80% of men are found to have factors contributing to osteoporosis when undergoing an evaluation for underlying causes of the disease (2, 3). It is important to exclude secondary causes of osteoporosis as the treatment of these patients may differ, and its response may be limited if the underlying disorder is unrecognized and left untreated. For example, alendronate has a reduced efficacy in post-menopausal women with osteoporosis receiving thyroid-stimulating hormone (TSH)-suppressive doses of levothyroxine (l-T4) for the management of differentiated carcinoma of the thyroid (4). Similarly, gain in bone mineral density (BMD) is greater in post-menopausal women with hypercalciuria receiving a bisphosphonate along with indapamide to treat the hypercalciuria, than in patients receiving a bisphosphonate alone (5). Some of the common conditions associated with osteoporosis and increased risk of fractures are listed in Table 1. Certain drugs may also increase the risk of bone loss and fractures (Table 2), and bone health issues should be considered when prescribing these medications.

Table 1

Common diseases associated with osteoporosis.

Endocrine disorders
 Glucocorticoid-induced osteoporosis
 Diabetes mellitus
 Growth hormone deficiency and acromegaly
Gastrointestinal, hepatic and nutritional disorders
 Celiac disease
 Inflammatory bowel disease
 Gastric bypass surgery
 Anorexia nervosa
 Hemochromatosis and chronic liver diseases
Hematological disorders
 Monoclonal gammopathy of uncertain significance
 Multiple myeloma
 Systemic mastocytosis
 Beta thalassemia major
Renal disorders
 Idiopathic hypercalciuria
 Renal tubular acidosis
 Chronic kidney disease
Autoimmune disorders
 Rheumatoid arthritis
 Systemic lupus erythematosus
 Ankylosing spondylitis
 Multiple sclerosis
Table 2

Drug-induced osteoporosis.

Hormones and drugs with actions on the endocrine system
 Thyroid hormone
 Hypogonadism-inducing agents
 Aromatase inhibitors
 Medroxyprogesterone acetate
 GnRH agonists
Drugs with actions on the CNS
Drugs with actions on the immune system
 Calcineurin inhibitors
 Antiretroviral therapy
Anticoagulants; heparin
Diuretics: loop diuretics
Drugs with actions on the gastrointestinal tract
 Proton pump inhibitors

In this review, common causes of secondary osteoporosis and the underlying mechanisms of bone loss are presented. Genetic conditions that directly affect the skeleton, such as osteogenesis imperfecta, are associated with an increased risk of fractures and were recently reviewed in the journal, so they are not included in this review (6).

Endocrine causes of osteoporosis

Glucocorticoid-induced osteoporosis

Glucocorticoid-induced osteoporosis (GIO) is the most common form of secondary osteoporosis. Although endogenous hypercortisolism or Cushing's syndrome can be associated with bone loss, most of the patients suffering from GIO receive glucocorticoids for the treatment of a variety of diseases. A fundamental point to recognize is that glucocorticoids are often administered to patients with inflammatory and autoimmune disorders, and the underlying disease itself is frequently a cause of osteoporosis. Inflammatory bowel disease (IBD), rheumatoid arthritis (RA) and systemic lupus erythematosus (SLE) are associated with bone loss due to the systemic release of inflammatory cytokines with significant effects on bone remodeling. In fact, older women with high inflammatory burden are at an increased risk of hip fractures (7). Glucocorticoid administration itself is associated with side effects, including osteoporosis and their use is characterized by a significant increase in fractures. Approximately 250 000 men and women take long-term glucocorticoids in the UK and are at a significantly increased risk of osteoporotic fractures and hip fractures independent of prior fractures (8). The cumulative dose and duration of glucocorticoid exposure are determinants of the risk of fractures (9). However, prolonged exposure to doses of prednisolone as low as 2.5–5 mg daily are associated with an increased risk of hip and vertebral fractures; a risk that increases with higher doses of prednisolone (10). Pre-menopausal women are protected, to some extent, from the detrimental actions of glucocorticoids on the skeleton, and post-menopausal women are at a greater risk of fractures. Indeed, no fractures were observed in pre-menopausal women in the alendronate intervention trial in either placebo or alendronate treated subjects with GIO (11). This may be because pre-menopausal women have a healthier skeleton than post-menopausal women when started on glucocorticoids. Importantly, the risk of fractures following intermittent glucocorticoids is modest and the risk decreases after discontinuation of glucocorticoids, suggesting that short-term exposure to these steroids does not have major detrimental effects on bone structure (12). Inhaled glucocorticoids have minimal effects on bone metabolism due to their limited absorption (13).

Glucocorticoids have both direct and indirect effects on bone metabolism (14). By blocking vitamin D actions on calcium absorption, glucocorticoids may cause transient decreases in serum calcium and modest increases in the serum levels of parathyroid hormone (PTH). However, secondary hyperparathyroidism does not explain the bone loss following glucocorticoid treatment, and serum PTH levels are not in the hyperparathyroid range. During the period of initial exposure to glucocorticoids, there is an increase in bone resorption, which is most likely responsible for a rapid decline in BMD and initial increase in fracture risk. The mechanism by which glucocorticoids enhance bone resorption is an induction of receptor activator of NF-KB ligand (RANKL) and of macrophage stimulating factor (MCSF), both required for osteoclastogenesis, and a decrease in the RANKL decoy receptor osteoprotegerin (OPG) (15). Glucocorticoids, however, may inhibit osteoclast bone degrading capacity and disruption of glucocorticoid receptor in osteoclast precursors not only reverses these effects of glucocorticoids but also protects from their inhibitory effects on bone formation (16). As skeletal exposure to glucocorticoids continues, bone formation is suppressed and a state of decreased bone remodeling ensues. The mechanism involves direct effects of glucocorticoids on cells of the osteoblastic lineage; cell number decreases because of an inhibition of osteoblast precursor cell replication and differentiation, and an increase in osteoblast apoptosis. In the presence of glucocorticoids, mesenchymal cells are directed toward cells of the adipocytic lineage instead of toward osteoblasts (17). A mechanism by which glucocorticoids inhibit osteoblast cell differentiation is by opposing Wnt/β-catenin, and Wnt is a critical regulator of osteoblastogenesis and also an inhibitor of bone resorption (18, 19).

Glucocorticoids not only deplete the osteoblast cell population but also inhibit mature osteoblast function directly, and indirectly by suppressing the expression of insulin-like growth factor 1 (IGF1) by skeletal cells (20, 21, 22). Glucocorticoids decrease the secretion of growth hormone (GH) and may alter the GH/IGF1 axis and also inhibit the release of gonadotropins, and as a result decrease estrogen and testosterone production and cause hypogonadism and bone loss (23, 24). Osteocytes are mechanosensors that play a role in the repair of bone microdamage, and glucocorticoids induce their apoptosis (22).

11 Beta-hydroxysteroid dehydrogenases (11β-HSDs) are isoenzymes that catalyze the interconversion of hormonally active cortisol and inactive cortisone and can regulate glucocorticoid activity (25). 11β-HSD1, a low affinity NADPH-dependent enzyme, is a bidirectional dehydrogenase/reductase, although it displays primarily reductase activity, converting cortisone to cortisol (26). 11β-HSD1 is widely expressed in glucocorticoid target tissues, including adult bone, and activates inactive glucocorticoids to facilitate glucocorticoid action. Glucocorticoids enhance 11β-HSD1 mRNA expression and activity in human osteoblasts and non-skeletal cells, and the effect could result in the amplification of the cellular actions of glucocorticoids in osteoblasts, where the enzyme may act as a positive autoregulator of glucocorticoid action (27, 28, 29).

Fractures in the context of GIO occur frequently at sites rich in cancellous bone, such as vertebrae and femoral neck, and vertebral fractures are often asymptomatic (30). A direct relationship between BMD and fracture risk in GIO has not been established, and post-menopausal osteoporosis diagnostic criteria should not be applied to GIO. Indeed, fractures in the context of glucocorticoid exposure occur at higher BMD values than in post-menopausal osteoporosis, and therapeutic intervention should be considered at T-scores that are in the osteopenic range (31, 32). FRAX, a computer algorithm that calculates the 10-year probability of major osteoporotic and hip fractures, can be used to estimate fracture risk in GIO. However, its value is limited by the fact that the use of glucocorticoids is entered as a dichotomous risk factor without consideration to the dose or length of exposure to glucocorticoids. Ways to correct for this limitation have been proposed, but their value has not been fully tested (8, 33). Changes in biochemical markers of bone turnover are dependent on the stage of the disease and have limited diagnostic value. Following high dose administration of i.v. glucocorticoids, there is an early and persistent decrease in bone formation markers and a transient increase in bone resorption markers (34). Markers of bone formation such as serum osteocalcin also are suppressed in patients with endogenous Cushing's syndrome (35).

Administration of supplemental calcium and vitamin D, use of the minimal effective dose of gucocorticoids to control the underlying disease and lifestyle changes are all recommended in the management of GIO (31, 36). Glucocorticoids oppose the effects of vitamin D on calcium absorption, although the mechanisms are poorly understood. In vitro studies have shown that glucocorticoids can increase or decrease the expression of the vitamin D receptor, suggesting that other mechanisms may be responsible for these effects of glucocorticoids (37, 38). Subjects receiving glucocorticoids may require higher doses of vitamin D3 daily, but calcium and vitamin D supplementation alone may not prevent the detrimental effects of glucocorticoids on the skeleton (39).

Bisphosphonates are indicated for the prevention and treatment of GIO, and teriparatide is indicated for the treatment of the disease. Alendronate, risedronate and zoledronic acid prevent the early decline in BMD in GIO and increase BMD in patients with established GIO (11, 40). However, evidence of reduction in vertebral fractures is only available for risedronate and benefit at non-vertebral sites was demonstrated only in observational studies (41). This is in part due to the fact that incidence of fractures was not a primary endpoint of the studies testing drug efficacy in GIO (33). Bisphosphonates are beneficial since there is a period of increased bone resorption following exposure to glucocorticoids, and their administration serves to stabilize BMD. Their use in pre-menopausal women needs to be considered with caution, since they cross the placenta and may affect embryonic skeletal development. Teriparatide is an option for the treatment of GIO since glucocorticoids have pronounced negative effects on osteoblast differentiation and function, and teriparatide is more effective than alendronate in increasing BMD at the lumbar spine and total hip (42). Although not a trial endpoint, subjects in the teriparatide arm had substantially less fractures than subjects in the alendronate arm (42).

Hyperthyroidism, thyroid hormone replacement and suppressive therapy

Euthyroidism is essential for normal skeletal development and linear growth and for the attainment of peak bone mass in early adulthood. Thyroid hormone deficiency in children results in impaired skeletal development and delayed bone age, while hyperthyroidism is associated with accelerated skeletal development and advanced bone age (43). Both hyperthyroidism and hypothyroidism have been associated with osteoporosis and increased risk of fractures. Thyrotoxicosis results in an increase in bone turnover, shortening of the bone remodeling cycle and uncoupling of bone remodeling, and can cause a loss of up to 10% of mineralized bone per remodeling cycle, while hypothyroidism can lengthen the bone remodeling cycle (44).

Suppressed serum TSH and a history of hyperthyroidism are associated with an increased risk of hip and vertebral fractures (45, 46, 47). In addition, ongoing therapy with thyroid hormone replacement is inversely correlated with BMD and increases the risk of fractures even in the presence of euthyroidism (48). TSH also has direct effects on the osteoclasts and osteoblasts (49). However, low BMD in peri-menopausal women seems to be dependent on serum levels of free thyroid hormones (50).

Several factors including age and sex of the patient, duration of treatment with T4 and the presence of additional predisposing factors may influence the impact of thyroid status on the skeleton, with older post-menopausal women being at the greatest risk for bone loss (45, 51, 52, 53). There are no specific guidelines for the prevention of bone loss secondary to hyperthyroidism. Supplemental calcium and vitamin D should be administered; and because thyroid hormone increases bone remodeling, antiresorptive agents may be considered in post-menopausal women at an increased risk of fractures.

Hypogonadism and agents inducing hypogonadism

Hypogonadism is associated with bone loss in men and women. It is the main underlying physiological change in post-menopausal women associated with low BMD and idiopathic osteoporosis. Premature menopause and medications, such as aromatase inhibitors and gonadotropin-releasing hormone (GnRH) analogs which cause hypogonadism, are associated with low BMD and increased risk of fractures. The effect of estrogen deficiency related to menopause and its contribution to post-menopausal osteoporosis is beyond the scope of this review which focuses on secondary causes of the disease. Hypogonadism is the most common cause of osteoporosis in men and is present in up to 20% of men with symptomatic vertebral fractures and 50% of elderly men with hip fractures (54). Both primary hypogonadism and testosterone deficiency due to androgen deprivation therapy are associated with an increased risk of osteoporosis and fractures (54, 55, 56). Men with osteoporosis may present either with symptomatic or asymptomatic hypogonadism and low serum levels of free testosterone.

Androgens have dual effects on the skeleton; a direct effect through activation of the androgen receptor and an indirect effect, following their aromatization to estrogen, which then signals through activation of estrogen receptor alpha (ERα). In men, the testes account for ∼15% of circulating estrogens and the remaining 85% is derived from peripheral aromatization of circulating androgen precursors in peripheral tissues. Both androgens and estrogens influence bone resorption in men independently. Androgens also regulate bone formation in men as shown in a study of young transgender men where testosterone treatment resulted in a significant increase in bone formation markers and hip BMD with no significant changes in estradiol (E2) levels, suggesting that this effect of testosterone was independent of its aromatization to estrogens (57).

Testosterone and 5α-dihydrotestosterone regulate gene expression in osteoblasts and inhibit the resorptive capacity of isolated human and murine osteoclasts in vitro (58, 59). In addition, sex hormones influence the secretion of a variety of cytokines and growth factors from skeletal cells including MCSF, and the proinflammatory molecules interleukin 1 (IL1), IL6 and tumor necrosis factor alpha (TNFα), RANKL and OPG, which contribute to the effects of androgens on bone remodeling (58, 60, 61).

Androgen replacement therapy in hypogonadal men with osteoporosis results in an increase in BMD, and the greatest increase is typically observed during the first year of therapy (62). As a consequence, androgen replacement therapy may be used in symptomatic patients to control hypogonadal symptoms as well as the skeletal manifestations of hypogonadism. Bisphosphonates are effective in the treatment of male osteoporotic patients with or without hypogonadism (63). Higher RANKL levels have been reported with low testosterone levels, and the RANKL antibody denosumab is effective in the treatment of men with osteoporosis (64, 65).

Aromatase inhibitors

Aromatase inhibitors are used in the management of ER positive breast cancer and act by inhibiting the aromatization of androgens, resulting in lower estrogen levels and bone loss. Anastrazole and letrozole are non-steroidal aromatase inhibitors while exemestane is a steroid that binds aromatase irreversibly (66). Aromatase inhibitors are effective in the treatment of infertility in polycystic ovarium syndrome, although no adverse skeletal effects should be expected following the short duration of exposure for infertility treatment of pre-menopausal women (66, 67). In most trials, the effectiveness and adverse event profile of aromatase inhibitors are compared to tamoxifen. Aromatase inhibitors increase biochemical markers of bone turnover, decrease BMD and increase the risk of fractures when compared to tamoxifen (68, 69, 70). Bisphosphonates and denosumab increase BMD in patients treated with aromatase inhibitors, but data on fracture reduction efficacy following antiresorptive therapy are limited.

Medroxyprogesterone acetate

Depot medroxyprogesterone acetate is an injectable progestogen used as a contraceptive agent and in the treatment of endometriosis. Medroxyprogesterone acetate inhibits gonadotropin secretion, suppressing ovulation and estrogen production by the ovary; and as a consequence, it causes a decrease in BMD. This is reversible following the discontinuation of medroxyprogesterone, and whether there is an increase in the risk of fractures is not established (71).

GnRH agonists

GnRH agonists are analogs with increased receptor affinity or prolonged half-lives that lead to persistent activation of GnRH receptors and suppression of gonadotropin secretion, with a consequent decrease in ovarian estrogen production. GnRH agonists are effective in the management of endometriosis and infertility in pre-menopausal women. By suppressing estrogen levels, they cause a decline in BMD; this is reversed following the discontinuation of the GnRH agonist (72, 73).

GnRH agonists are used as androgen deprivation therapy in patients with advanced carcinoma of the prostate. GnRH analogs reduce serum testosterone and E2 levels and increase bone turnover causing bone loss and increasing the risk of fractures (74, 75, 76, 77, 78). Men undergoing androgen deprivation therapy as part of the management of prostate cancer suffer 2–8% loss of vertebral BMD and 2–6% loss of BMD at the femoral neck compared to age-matched controls. The greatest decline in BMD occurs in the first year following initiation of androgen deprivation therapy and is associated with a significant decrease in sex steroid hormone levels and with an increase in bone turnover markers (79, 80).

Treatment is recommended for patients with osteoporotic fractures or with significant risk factors for fractures (81). Calcium and vitamin D are indicated. Bisphosphonates are effective in restoring BMD in patients suffering from bone loss secondary to androgen deprivation therapy, but data on fracture reduction are limited (80, 82). Selective ER modulators, such as raloxifene and toremifene, have beneficial effects on BMD, but are not approved for use in men (83, 84). Denosumab can prevent bone loss and reduces the incidence of vertebral fractures in men receiving androgen deprivation therapy (85).


Primary hyperparathyroidism (PHPT) is associated with an increase in the expression of RANKL by cells of the osteoblast lineage and an increase in osteoclast-mediated bone resorption. Although osteoblast activity and bone formation may increase, this is not sufficient to counteract the enhanced bone resorption. Bone turnover markers are normal or mildly elevated. PHPT is associated with cortical bone loss, so that BMD in the distal forearm and the hip are decreased, although trabecular bone may also be affected in PHPT (86, 87). There is an increased risk of vertebral, wrist, ankle, rib and pelvic fractures in PHPT (88, 89). Parathyroidectomy results in normalization of serum calcium levels and an increase in vertebral and femoral BMD. Bisphosphonates and hormone replacement therapy decrease bone remodeling and increase BMD in patients with PHPT (90). However, fracture data are not available and neither agent lowers serum calcium or PTH levels. Denosumab can correct the hypercalcemia of PHPT and has been used effectively in the treatment of hypercalcemia in patients with parathyroid carcinoma (91, 92, 93, 94, 95).

Diabetes mellitus

Diabetes mellitus (DM) and impaired glucose metabolism have a detrimental effect on bone metabolism, and both types 1 and 2 DM carry increased risk of fractures; the risk is higher with type 1 than with type 2 DM (96, 97). The risk for fractures is common to both men and women, and increases with the duration of the disease and the use of insulin (97, 98).

Although type 2 DM is associated with an increase in BMD associated with higher insulin and IGF1 levels, there is a higher occurrence of fractures in this population. As fractures occur at higher bone density in diabetics than in non-diabetic patients, BMD cannot predict the risk for fractures accurately in this population (99). In preliminary studies, trabecular bone score was shown to predict fracture risk in diabetic patients, but its use needs further validation (100). It is important to consider fracture prevention strategies at a relatively earlier stage in patients with DM. Diabetic patients should be counseled on adequate calcium and vitamin D intake, and glycemic control should be optimized to minimize microvascular complications, retinopathy and neuropathy, which may lead to an increased risk of falls and fractures.

Agents that have been used for the treatment of DM, such as thiazolidinediones, are associated with an increased risk of fractures. Thiazolidinediones are insulin-sensitizing drugs that activate peroxisome proliferator-activated receptor gamma, a transcription factor that controls mesenchymal cell allocation by increasing adipogenesis and decreasing osteoblastogenesis (17, 101, 102). Thiazolidinediones also promote osteoclast differentiation and bone resorption, and patients may experience bone loss and low BMD (103, 104, 105).

GH deficiency and acromegaly

GH plays an important role in linear growth and attainment of appropriate height and peak bone mass during childhood and adolescence (106). GH increases bone formation by interacting directly with GH receptors on osteoblasts and via locally produced IGF1. GH deficiency (GHD) has been associated with delayed skeletal maturation and low BMD in patients with isolated GHD as well as in patients with multiple pituitary deficiencies, predominantly through decreased bone formation (107). Several studies have reported a higher prevalence of osteoporosis and a higher fracture rate in adult onset GHD and morphometric vertebral fractures have been reported in up to half of patients with GHD (108). Treatment of GHD with GH increases vertebral and femoral BMD over time. Recombinant human GH has a biphasic effect on bone; an initial phase associated with an increase in bone resorption and a decrease in BMD and a second phase characterized by an increase in bone formation and in BMD, usually after 6–12 months of treatment (109).

Acromegaly is associated with increased bone remodeling, and patients with acromegaly have a significantly higher prevalence of vertebral fractures, which correlate with the duration of the disease and serum IGF1 levels (109, 110). Radiological vertebral fractures occur in as many as a third of the patients with acromegaly without changes in BMD, so that BMD has limited value in the assessment of these patients. A formal vertebral morphometric analysis is recommended in patients with acromegaly to evaluate for the presence of occult vertebral fractures. Although treatment of acromegaly improves bone health, the increased risk of fractures may persist in patients with hypogonadism and with prior history of vertebral fractures (111).

Gastrointestinal, nutritional and hepatic causes of osteoporosis

Celiac disease

Celiac disease is associated with bone loss and an increased risk of fractures. About one-third of patients with celiac disease have osteoporosis, with men being more severely affected than women (112). In children, celiac disease is associated with delayed bone growth and puberty due to nutritional deficiency and malabsorption (113). Both symptomatic and asymptomatic patients with celiac disease have low BMD, and tissue transglutaminase antibody IgA seropositivity status is associated with osteoporosis and increased risk of hip fractures (114, 115). Decreased calcium absorption, the consequent secondary hyperparathyroidism and an increase in the levels of inflammatory cytokines, including TNFα, IL1 and IL6, may be responsible for an increase in bone resorption (116, 117). Malabsorption of micronutrients may contribute to altered bone metabolism.

Institution and long-term adherence to a gluten free diet improves BMD in children and adults, although BMD is seldom normalized. There is no agreement on when to obtain BMDs in patients with celiac disease, and its value to predict fracture risk has also not been established in this disease (116). There are no longitudinal studies evaluating the efficacy of bisphosphonates in celiac disease, although their use is common when patients manifest osteoporosis. Gluten free diet and optimal calcium and vitamin D supplementation should be initiated.

Inflammatory bowel disease

IBD is comprised by Crohn's disease and ulcerative colitis, and both are associated with osteoporosis. The prevalence of osteoporosis varies and is related to the severity of the IBD and related comorbid conditions. Osteoporosis can present in more than one-third of patients with established IBD and is associated with an increased frequency of vertebral and hip fractures (118, 119). Children with IBD may fail to achieve peak bone mass (120). The mechanisms responsible for the bone loss in IBD include disease-related inflammatory activity and treatment-related side effects, including glucocorticoid therapy, nutritional deficiencies, leading to low BMI and contributing to hypogonadism. Disease severity correlates with fracture risk even after adjusting for corticosteroid use. The risk of fractures is higher in Crohn's disease than in ulcerative colitis (121).

Serum levels of the pro-inflammatory cytokines TNFα, IL1β, IL6 and IL17 are increased in IBD (122, 123). These cytokines enhance osteoclastogenesis and bone resorption and inactivation of the TNF type 1 receptor abrogates the effect of TNFα on osteoclastogenesis in bone marrow macrophage cultures, suggesting that increased TNF production contributes to the bone loss found in patients with IBD (124, 125). This is substantiated by studies demonstrating that treatment of patients with IBD with the anti-TNFα antibody infliximab improves vertebral BMD (126).

Ongoing treatment with glucocorticoids contributes to the bone loss and increased risk of fractures (127). In prospective studies in children and adolescents, greater lifetime prednisolone exposure correlates with lower lumbar spine BMD, reflecting both a more active disease and the effects of glucocorticoids (120, 128, 129).

Vitamin D deficiency is common in patients with Crohn's disease, particularly in patients with prior ileal resection (130, 131). Lower calcium intake and absorption, as well as vitamin K nutritional deficiencies, may contribute to the bone loss (132).

Patients with IBD should be evaluated for osteoporosis, and calcium and vitamin D intake need to be optimized. Patients should be maintained on the lowest dose of glucocorticoids needed to maintain the disease in remission, and antiresorptive therapy should be used in patients with osteoporosis and fractures.

Gastric bypass surgery

In the past decade, there has been an increase in the number of cases undergoing bariatric surgery, a surgical procedure associated with bone loss. The surgical technique used may play a role in the bone loss. Osteoporosis and osteomalacia, resulting from defective mineralization are long-term complications of gastrectomy. Surgical techniques that result in malabsorption lead to malnutrition, calcium and vitamin D deficiency with consequent secondary hyperparathyroidism, all of which contribute to bone loss. Vitamin D supplementation is recommended in these patients.

Although there is a higher prevalence of osteopenia and vertebral fractures in patients following gastric bypass surgery, the exact risk for osteoporosis and fractures is unknown (133). Patients undergoing bypass surgery have low BMD, but there are technical limitations of the dual energy X-ray absorptiometry technology in this population due to the weight of the patients. There is a lack of long-term prospective data on BMD in this patient population. Therefore, it is difficult to determine the impact of the surgery on BMD.

Eating disorders

Anorexia nervosa is associated with significant weight loss, hypogonadotropic hypogonadism with amenorrhea, low BMD and an increased risk of fractures (134). Both cortical and trabecular bone are affected, bone accrual during growth is slow and peak bone mass is low. Serum markers of bone formation are suppressed and markers of bone resorption are increased suggesting that bone formation is uncoupled from bone resorption. The best strategy to improve bone health is to regain weight and ovarian function. Oral estrogen–progesterone combinations are not effective in increasing BMD in adults or adolescents with anorexia nervosa. However, physiological estrogen replacement as transdermal E2 with cyclic progesterone increase bone accrual rates in adolescents with anorexia nervosa, leading to maintenance of BMD. Risedronate also increases vertebral and hip BMD, but bisphosphonates should be used with great caution in women of reproductive age, given the long half-life of bisphosphonates and potential for teratogenicity (135).

Hemochromatosis and other chronic liver diseases

Hemochromatosis is a relatively common genetic disorder, which results in iron overload and iron deposition in several tissues including the liver, pancreas and the pituitary gland. Osteoporosis can be prevalent in one-quarter to one-third of patients with hemochromatosis (136). Several factors contribute to osteoporosis in patients with hemochromatosis including secondary hypogonadism, liver failure and vitamin D deficiency. Iron also has direct toxic effects on the skeleton, resulting in increased bone resorption and decreased bone formation (137). It is not known whether the treatment of hemochromatosis restores skeletal architecture or BMD. At present, guidelines for post-menopausal and male osteoporosis are recommended for the management of osteoporosis in these patients.

Metabolic disturbances of bone are associated with chronic liver diseases particularly those affected by cholestasis, such as primary biliary cirrhosis and primary sclerosing cholangitis (138). The mechanisms responsible for osteoporosis in chronic liver disease are complex and include malnutrition, glucocorticoid use, vitamin D deficiency and hypogonadism, as well as complications that follow liver transplantation (138). Patients with end stage liver disease are at significant risk of bone loss and 40–60% present with BMD in the osteopenic or osteoporotic range and nearly 25% suffer fragility fractures of the spine or ribs (139). The prevalence of osteoporosis/osteopenia in patients with primary biliary cirrhosis is 20–40% depending on the stage of the disease; and about 10–20% of this patient population presents with fragility fractures (140, 141). Management consists of decreasing risk factors and possible anti-fracture intervention, although data on the efficacy of therapeutic intervention in chronic liver disease are limited. There is evidence that bisphosphonates increase BMD in patients with chronic liver disease, but the number of patients studied is too limited to demonstrate anti-fracture efficacy (142, 143, 144).

Hepatitis C is a common form of hepatitis, and 5–20% of patients will progress to develop cirrhosis (145). However, even in the absence of cirrhosis, patients with chronic hepatitis C infection present with osteopenia or even osteoporosis, as defined by low BMD (146, 147). The mechanisms responsible have not been established, although chronic inflammation may play a role. There is no apparent increase in the incidence of fractures in patients with hepatitis C infection without cirrhosis and some studies have reported normal BMD in well-nourished patients with hepatitis C infection (147, 148). However, hepatitis C infection may contribute to the risk of fractures among patients with human immunodeficiency virus (HIV) infections (149).

Hematological disorders and osteoporosis

Monoclonal gammopathy of uncertain significance

Monoclonal gammopathy of uncertain significance (MGUS) is characterized by an overproduction of monoclonal protein and an imbalance in bone remodeling that leads to diffuse bone loss. MGUS and multiple myeloma are common causes of osteoporosis associated with fragility fractures. MGUS is characterized by a plasma cell content of <10% in the bone marrow, a monoclonal (M) protein spike of ≤30 g/l and no end organ damage (absence of skeletal lytic lesions, hypercalcemia, renal insufficiency, anemia and bone lesions) (150).

MGUS patients are at increased risk for osteoporosis, vertebral and hip fractures, and a higher prevalence of MGUS is found in patients older than 50 years of age who present with hip fractures (151, 152, 153, 154, 155). Typically, MGUS is associated with increased bone resorption and reduced bone formation, and the Wnt antagonists, Dickkopf-1 (Dkk-1), soluble frizzled related proteins 2 and 3 and sclerostin have been associated with skeletal disease and bone loss (156, 157, 158, 159, 160, 161). Wnt signaling enhances osteoblast differentiation and suppresses osteoclastogenesis, thereby contributing to the maintenance of bone mass. Interference of Wnt signaling by Wnt antagonists and an increase in RANKL by inflammatory cytokines contribute to the bone loss in MGUS (162).

Current guidelines do not recommend the use of bisphosphonates in patients with MGUS. Alendronate and zoledronic acid cause a significant increase in BMD in patients with MGUS, but no data are available on fracture risk reduction by these agents (163, 164).

Multiple myeloma

Multiple myeloma is a plasma cell disorder, characterized by bone marrow infiltration with clonal plasma cells, production of monoclonal immunoglobulin and end organ damage including lytic lesions in bone with hypercalcemia, renal impairment and bone marrow involvement with anemia. The clonal proliferation results in classic lytic lesions and reduced bone mass due to increased bone resorption and decreased bone formation. Myeloma cells achieve these effects through the production of several cytokines, such as IL6 and IL7, leading to an increase in RANKL production by bone marrow stromal cells and an increased degradation of OPG. This results in an increased RANKL/OPG ratio, the differentiation of osteoclast precursors and enhanced bone resorption (165, 166). The expression of the Wnt antagonists, Dkk-1 and secreted frizzled protein-2, are increased in myeloma explaining the decrease in osteoblastogenesis through an inhibition of Wnt signaling (159, 167). The net effect of these changes is decreased bone formation and increased bone resorption, leading to osteoporosis. Bisphosphonates and targeted therapies for myeloma help reduce osteolysis and the risk of fractures (168). Denosumab may be an alternative for the prevention of skeletal-related complications in patients with multiple myeloma with impaired renal function since the drug is not metabolized or excreted by the kidneys (169).

Systemic mastocytosis

Systemic mastocytosis is a hematologic condition commonly associated with reduced bone mass and osteoporosis, and is present in ∼9% of bone biopsies from men with idiopathic osteoporosis (170). Vertebral compression fractures are relatively common in men and women with systemic mastocytosis (171). Generally, there is greater bone loss at the spine than at the hip. Spine X-rays and BMD are warranted in patients with or without skin involvement and particularly in males. Diagnosis is based on an elevated serum tryptase level and increased 24-h urine excretion of N-methylhistamine, but diagnosis requires histologic confirmation demonstrating mast cell infiltration in the bone marrow as well as the finding of c-KIT mutations. Systemic mastocytosis is frequently associated with mutations of the KIT receptor tyrosine kinase (c-KIT), leading to the constitutive activation of c-KIT in most cells (172). Therefore, genetic analysis of c-KIT in patients with mastocytosis can be of diagnostic value.

The proximity of the mast cell to bone remodeling surfaces and the production of chemical mediators and cytokines, such as histamine, prostaglandins, leukotrienes and cytokines (IL1, IL3 and IL6) contribute to the skeletal pathology (173). Patients have higher levels of IL6 and RANKL leading to an increase in bone resorption and higher levels of serum sclerostin possibly contributing to a decrease in Wnt signaling and bone formation (174). However, the mechanisms responsible for osteoporosis and increased fracture risk need further study so that targeted therapies can be evaluated. Bisphosphonates such as alendronate, pamidronate and zoledronic acid can improve BMD and decrease bone resorption in patients with mastocytosis (175, 176).

Beta thalassemia major

Thalassemias comprise a group of inherited disorders of hemoglobin synthesis, and mutations in the BGLOBIN gene lead to defective β globin production and anemia. The care of these patients with transfusions and iron chelation has improved patient survival; and as a result, thalassemia major is now associated with osteoporosis. This is a prominent cause of morbidity among affected patients (177). The etiology of the osteoporosis is multifactorial. Enhanced bone resorption and remodeling and suppressed osteoblastic function with bone loss and increase risk of fractures occur in patients affected with thalassemia major. Bisphosphonates decrease bone remodeling and increase BMD in patients with thalassemia major, although it is not known whether fracture risk is decreased with bisphosphonate therapy (178).

Renal causes of secondary osteoporosis

Idiopathic hypercalciuria

Idiopathic hypercalciuria is associated with a 24-h urinary calcium excretion in excess of 4 mg/kg in women or of 4.5 mg/kg in men without an underlying cause. Idiopathic hypercalciuria is associated with low BMD and an increased prevalence of fractures, as calcium excretion is higher than absorption, resulting in a net calcium loss. In the Third National Health and Nutrition Examination Survey, men with a history of kidney stones were found to have lower femoral neck BMD, elevated markers of bone turnover and an increased risk of wrist and vertebral fractures (179, 180, 181). IL1, IL6 and TNFα are increased in hypercalciuria and are associated with elevated markers of bone turnover (182).

The bone loss associated with idiopathic hypercalciuria may be caused by a primary disorder of bone formation or resorption, or may be secondary to the abnormal renal handling of calcium and sodium. The underlying mechanisms of the bone loss are not entirely known. Idiopathic hypercalciuria is characterized by increased intestinal calcium absorption, increased bone resorption and decreased renal tubular calcium reabsorption (180). Higher circulating 1,25-dihydroxyvitamin D3 levels and an increased expression of the vitamin D receptor in monoctyes occur in the majority of patients with hypercalciuria and renal stones (183, 184). Studies in genetic hypercalciuric stone forming rats show a decrease in urinary calcium excretion with bisphosphonates, suggesting that the increase in urinary calcium is secondary to enhanced bone resorption (185).

Patients with idiopathic hypercalciuria are managed with thiazides to reduce calcium excretion and bisphosphonates to inhibit bone resorption when osteoporosis is present. Bisphosphonates also reduce urinary calcium excretion in normal and hypercalciuric patients (186, 187). Thiazides act by stimulating calcium reabsorption in the distal convoluted tubule and also stimulate osteoblast differentiation with a reduction in stone recurrence rate, osteoporotic fractures and an increase in BMD (186, 188, 189, 190, 191, 192, 193, 194, 195).

Renal tubular acidosis

Renal tubular acidosis (RTA) is a metabolic acidosis caused by either a reduced capacity of the proximal tubule of the kidney to reabsorb the filtered bicarbonate load (proximal) or a reduced capacity of the distal renal tubule to acidify the urine maximally (196). When the hydrogen ion load is greater than the normal daily acid load, bone buffers the hydrogen ions. This may result in a spectrum of metabolic bone disorders ranging from osteomalacia (with proximal RTA) to osteoporosis (with distal RTA) and fractures. Defective renal acidification may lead to an osteoblast-mediated activation of osteoclasts and a compensatory mobilization of alkali and calcium from bone resulting in bone loss (197, 198). Calcium reabsorption in the cortical collecting tubule is also reduced in RTA resulting in a negative renal calcium balance. Electrolyte abnormalities including increased chloride (>110 mEq/l) and low bicarbonate (<18 mEq/l) raise suspicion for RTA, and patients should be evaluated with arterial blood gases and a determination of urinary pH.

RANKL, TNFα and prostaglandins have been implicated in acid-induced, cell-mediated net calcium efflux from bone (196, 199). Simultaneous inhibition of RANKL and TNFα reduce the acid-induced, cell-mediated net calcium efflux from bone in in vitro studies. There are no studies evaluating the role of antiresorptive therapy in RTA.

Chronic kidney disease

The etiology of osteoporosis and fractures in chronic kidney disease (CKD) is multifactorial. The diagnosis of osteoporosis can be made in stages 1–3 CKD only on the basis of a low BMD or a fragility fracture, in the absence of concomitant metabolic abnormalities that suggest the presence of CKD–mineral and bone disorders (200). The metabolic bone disorder encompasses systemic derangements in bone mineralization and turnover associated with systemic vascular calcifications. CKD patients are more likely to fracture because they are at an increased risk for falls from muscle weakness and impaired balance secondary to poor nutrition, inactivity, myopathy and peripheral neuropathy (201). A detailed discussion of renal osteodystrophy is beyond the scope of this review, and reviews of this disorder have recently been published (202).

Autoimmune disorders and osteoporosis

Rheumatoid arthritis

RA is a chronic inflammatory disease associated with marginal joint erosions, periarticular osteopenia and systemic osteoporosis. Patients with RA have a two- to threefold increased risk of hip and vertebral fractures (203, 204). Underlying disease activity and ongoing use of glucocorticoids can contribute to bone loss and risk for fractures. Several cytokines involved in RA-associated inflammation such as IL1, IL6 and TNFα can promote osteoclastic activity. Increased RANKL/OPG ratio, elevated bone turnover markers and sedimentation rate are predicators of rapid and persistent bone loss and articular erosions in patients with RA (205).

Anti-TNF antibodies, used in the treatment of the inflammatory process in RA decrease the systemic bone loss (206). Increased levels of Wnt antagonists, DKK-1 and sclerostin also contribute to the decreased bone formation leading to bone loss; IL6 inhibitors reverse these changes (207, 208).

Bisphosphonates are used in the treatment of osteoporosis associated with RA. In a meta-analysis evaluating the use of bisphosphonates on BMD and prevention of vertebral and non-vertebral fractures, a significant reduction in incident vertebral fractures was observed after 18 months of bisphosphonates when used for prevention and after 36 months when used for treatment with preservation of BMD (209).

Systemic lupus erythematosus

Low BMD occurs in up to 50% of female patients with SLE in the premenopausal age. Race, glucocorticoid intake and cumulative corticosteroid exposure and use of anticoagulants such as heparin are predictors of reduced BMD and osteoporosis (210, 211). Other factors such as immunosuppressive drugs, limited physical activity and vitamin D deficiency contribute to the reduced bone mass. A 10–12% increased risk of fractures is reported in SLE patients. Osteoclast-inducing inflammatory cytokines such as IL6, soluble IL6 receptor, IL1 and TNFα are elevated in SLE and contribute to the bone loss. There is higher incidence of vitamin D deficiency in patients with SLE, possibly because patients avoid sun exposure (212). Supplemental calcium and vitamin D intake should be encouraged. No specific guidelines exist to evaluate the risk of fracture. Management of GIO is discussed under endocrine causes of osteoporosis and bisphosphonates need to be used with caution in women of child bearing age.

Ankylosing spondylitis

Ankylosing spondylitis (AS) is a relatively common cause of chronic arthritis that predominantly affects men. Osteoporosis is seen in up to 25%, and osteopenia in up to 50%, of patients with AS with a higher incidence of both vertebral and non-vertebral fractures (213). The etiology of osteoporosis in patients with AS is multifactorial with systemic inflammation mediated by TNFα being the most important etiologic factor. Both RANKL induction possibly by TNFα and decrease in Wnt signaling have been implicated in the osteoclast-mediated bone resorption and the dysregulated bone formation in AS; TNFα inhibitors increase vertebral and hip BMD in this condition (214).

Multiple sclerosis

Multiple sclerosis is a chronic demyelinating neurologic condition frequently associated with significant disability and limitation of physical function. Osteoporosis occurs frequently in male and female patients with multiple sclerosis (215). Factors contributing to the low BMD in this population include vitamin D deficiency, glucocorticoid therapy and female gender. Degree of functional impairment, advanced age and duration of disease also contribute significantly to the low BMD found in these patients (216).

Studies comparing BMD between ambulatory patients with multiple sclerosis and healthy control subjects do not show significant differences in BMD (217). However, the level of disability in patients with multiple sclerosis, as measured by the Expanded Disability Status Score, correlates strongly with BMD at the lumbar spine and femoral neck (218).

Drug-induced osteoporosis

Drug-induced osteoporosis is common and can be caused by a variety of pharmacological agents, including those used for the treatment of endocrine, central nervous system (CNS), immune, cardiovascular and gastrointestinal disorders (Table 2).

Hormones and agents with actions on the endocrine system

Glucocorticoids, thyroid management therapy, and agents that induce hypogonadism and thiazolidinediones are discussed under the section ‘Endocrine causes of osteoporosis’.

Drugs with actions on the CNS

Drugs with direct effects on the CNS may have deleterious effects on the skeleton and include antidepressants and anticonvulsants.


The use of selective serotonin and norepinephrine reuptake inhibitors (SSRI/SNRI) for the treatment of depression can be associated with bone loss and an increase in the risk of fractures in individuals ≥50 years of age (219, 220). The mechanism remains controversial although subjects on SSRI and SNRI exhibit increased biochemical markers of bone remodeling (221). There are no specific guidelines for the prevention of the bone loss associated with SSRI/SNRIs, but screening for osteoporosis should be considered.


Patients with epilepsy are at an increased risk of fractures, and epilepsy itself or its treatment may influence the pathogenesis of fractures (222). Prospective studies have demonstrated a greater decrease in BMD in women ≥65 years of age on anticonvulsants, but many confounders exist in these studies and fractures also are associated with seizure activity (223, 224). Anticonvulsants may cause bone loss, but mechanisms responsible are not clear. There is accelerated vitamin D metabolism with certain anticonvulsants and this could lead to low 25-hydroxyvitamin D3 levels, high bone turnover and secondary hyperparathyroidism increasing the risk of fractures (225, 226). These changes correlate with the type of anticonvulsant and duration of the treatment. Moreover, there is limited understanding of the pathogenesis of this skeletal disorder so that diagnostic and therapeutic guidelines are not established. Antiepileptic agents known to induce the liver enzymes cytochrome P450 (CYP450) and uridine 5′-diphospho-glucuronosyltransferase include phenytoin, carbamazepine and phenobarbital, and at high doses topiramate (227). Use of these agents is associated with greater risk of fractures than use of non-enzyme inducing antiepileptic drugs (228). The mechanism involves hydroxylation of specific residues leading to the inactivation of 1,25-dihydroxyvitamin D3 (229). Similarly, polytherapy is associated with increased risk of fractures. A recent meta-analysis revealed a weak and non-significant association between the use of carbamazepine, valproic acid and gabapentine and fracture risk, and an increase risk of fractures with liver enzyme inducing drugs; namely, phenobarbiturate, topiramate and phenoytoin, of 40–80% (228). There has been an evolution in the management of seizure disorders and much of the evidence linking anticonvulsants to osteoporosis may not apply to newly developed drugs. Calcium, vitamin D supplementation and bisphosphonates were shown to have a beneficial effect on BMD in a small cohort of patients with epilepsy treated with anticonvulsants (230).

Anticholinergic drugs are frequently used in an elderly population for the treatment of a variety of conditions. In a large, population-based cohort of Canadians older than 50 years old, it was found that the use of anticholinergic medications appeared to be associated with an increased risk of falling and non-traumatic fractures in unadjusted analyses; however, after adjusting for confounding variables, there was not a significant association (231). Similarly, an accelerated decline in femoral neck BMD was seen in anticholinergic medication users, but this decline was accounted for by factors other than anti-cholinergic drug use and after covariate adjustment, no significant association was seen (231).

Drugs with actions on the immune system

Calcineurin inhibitors

Calcineurin inhibitors are immunosuppressants used in combination with glucocorticoids in patients undergoing organ transplantation and are associated with an increase in fracture risk (232). Calcineurin inhibitors decrease osteoclastogenesis and bone resorption in vitro (233). However, mouse models harboring either the activation or downregulation of calcineurin have generated conflicting results (234, 235). Calcineurin is a phosphatase that dephosphorylates nuclear factors of activated T-cells (NFATs). Nfatc1 is critical for osteoclast formation, and studies from our laboratory have revealed a direct inhibitory effect of Nfatc1 and Nfatc2 on osteoblastogenesis (236, 237, 238). Accordingly, the conditional inactivation of calcineurin in osteoblasts causes an increase in bone formation and a decrease in bone resorption (235). These observations do not explain the bone loss observed in humans treated with calcineurin inhibitors suggesting that the calcineurin/NFAT axis may not be responsible for the bone disease observed. Alternate targets of calcineurin or the concomitantly administered glucocorticoids may be responsible for the negative impact on skeletal homeostasis. Moreover, patients undergoing liver transplantation may present with osteoporosis, osteopenia or even vertebral fractures prior to transplantation (239). These observations would incriminate the underlying disease as the cause of the bone loss and fractures. Importantly, there is no association between BMD and fractures, highlighting the fact that post-menopausal osteoporosis diagnostic criteria are often not applicable to secondary osteoporosis. Since bone loss occurs prior to, or in the initial months of, immunosuppressive therapy, treatment should be instituted early. Calcium and vitamin D supplementation, and antiresorptive agents should be considered for the prevention of bone loss after transplantation (240).

Antiretrovial therapy

Antiretroviral agents are used to treat patients affected by infections with HIV. Cross-sectional studies have shown a three- to sevenfold increase in the risk of osteoporosis in patients with HIV infection and an increased risk of fractures in men and women with HIV infection (241). Patients with HIV infection, secondary to illicit drug use may be more exposed to trauma and consequent fractures than the general population. In addition, hepatitis C infection is a co-morbidity and may contribute to the risk of fractures in patients with HIV infections (149). A recent meta-analysis revealed that patients with HIV infection had a 1.35 incidence rate ratio of fragility fractures compared to controls (242). Predictors of fractures included older age, white race, low BMI, diabetes, liver disease, alcohol, tobacco and substance abuse. Chronic inflammatory conditions found in patients with HIV infection may be associated with increased RANKL expression and enhanced bone turnover. Antiretroviral agents also may cause bone loss although the mechanisms may vary (243). Tenofovir may alter renal function and cause secondary hyperparathyroidism with increased bone resorption. Non-nuclease RT inhibitors are associated with decreased serum levels of vitamin D and protease inhibitors may impair Wnt signaling and decrease osteoblastogenesis (241). Patients with HIV infection treated with tenofovir and certain protease inhibitors suffer greater losses of BMD but the relevance to fracture risk has not been determined. Guidelines for the management of antiretroviral therapy-induced bone loss are available from the European Aids Clinical Society.


Heparin is effective in the prevention and treatment of thromboembolic disorders. In vitro, heparin inhibits the differentiation and function of osteoblasts, and in vivo it decreases bone formation and increases bone resorption (244, 245). Heparin binds to OPG, the decoy receptor for RANKL, allowing RANKL to induce osteoclastogenesis (245, 246). About one-third of patients receiving long-term therapy with heparin display a decrease in BMD, but fractures seldom occur and are less frequent in patients on low-weight heparin than on unfractioned heparin (247, 248, 249).

Oral anticoagulants are often used to prevent or treat thromboembolic phenomena, and their effects on bone metabolism are controversial. Anticoagulants with vitamin K antagonist activity, such as warfarin, interfere with γ-carboxyglutamate formation, and consequently inhibit the accumulation of osteocalcin in the extracellular bone matrix (250). Although these are potentially negative effects, evidence that warfarin causes osteoporosis and fractures is insufficient (251, 252).


Loop diuretics are used in the management of congestive heart failure, which itself is associated with an increased risk of fragility fractures (253). Loop diuretics inhibit sodium and chloride reabsorption and inhibit calcium reabsorption, increasing its renal excretion (254). This results in increased bone turnover, decreased BMD and increased risk of osteoporotic fractures (255, 256). In contrast, thiazide diuretics cause calcium retention and an increase in serum calcium, and have modest or no effects on BMD or risk of fractures (257).

Drugs with actions on the gastrointestinal tract

Proton pump inhibitors

Proton pump inhibitors are commonly used in the management of disorders of the upper gastrointestinal tract. By increasing gastric pH, they may decrease calcium absorption and have negative effects on skeletal homeostasis. However, their impact on BMD or fragility fractures is controversial (258). A large prospective population based Canadian Multicenter Osteoporosis Study and the Women's Health Initiative Observational Study reported a modest association between long-term proton pump inhibitor use and risk of fragility fractures (259, 260, 261). Although the effect of proton pump inhibitors on fracture risk is modest, it is reasonable to treat these patients with supplemental calcium and vitamin D. A calcium supplement that can be absorbed in the presence of proton pump inhibitors should be considered.

Summary and recommendations

Patients with osteoporosis frequently have underlying causes for the disease and should be evaluated for their presence. A modification of factors contributing to the bone loss may reverse the bone loss and fracture risk. It is common to find suboptimal calcium and vitamin D intake in these patients, deficiencies that are easily correctable. Basic laboratory evaluation should be considered in all patients, and further testing can be done based on clues from history and initial evaluation to look for causes of bone loss (see Table 3). However, it is important to consider that the prevalence of secondary osteoporosis in patients with fractures is relatively low, particularly among post-menopausal women, and the laboratory costs may be substantial (262). When a drug is responsible for the bone loss, an effort should be made to use the lowest dose of the medication, as in the case of glucocorticoids so that the risk of bone loss and fracture risk is minimized.

Table 3

Evaluation of secondary causes of osteoporosis.

Initial workup
 Detailed history to identify risk factors for osteoporosis
 Evaluation of nutritional status, calcium and vitamin D intake
 Bone mineral density
 X-ray lumbar and thoracic spine (if height loss >1.5″)
 Complete blood count with differential
 Serum calcium, phosphate, creatinine, albumin and 25-hydroxyvitamin D
 Serum protein electrophoresis
 Serum total and free testosterone, FSH, LH and prolactin
 Bone resorption and formation markers
 24 h urine calcium, creatinine and sodium
Further workup
 Serum parathyroid hormone
 Serum thyroid-stimulating hormone
 Serum tissue transglutaminase antibodies
 Serum ferritin and liver function tests
 Serum electrolytes
 Serum tryptase levels
24 h urinary cortisol or dexamethasone suppression test
Nuclear bone scan (if bone turnover unexplained high)

The validity of BMD as a predictor of fractures has not been established in patients with secondary causes of osteoporosis. Fractures may occur at reasonably normal BMDs limiting the value of BMD in these patients. Treatment interventions may vary depending on the cause of bone loss since the underlying cause may influence the response to therapy. Whereas therapies directed to the treatment of osteoporosis may be effective, treating the underlying disease is essential for a successful outcome.

Declaration of interest

The authors declare that there is no conflict of interest that could be perceived as prejudicing the impartiality of the review.


This work was in part supported by grants AR063049 (E Canalis) and DK045227 (E Canalis) from the National Institutes of Health.


The authors thank Mary Yurczak for secretarial assistance.



Official journal of

European Society of Endocrinology




PainterSEKleerekoperMCamachoPM. Secondary osteoporosis: a review of the recent evidence. Endocrine Practice200612436445. (doi:10.4158/EP.12.4.436).


HarperKDWeberTJ. Secondary osteoporosis. Diagnostic considerations. Endocrinology and Metabolism Clinics of North America199827325348. (doi:10.1016/S0889-8529(05)70008-6).


NIH Consensus Development Panel on Osteoporosis Prevention, Diagnosis, and TherapyOsteoporosis prevention, diagnosis, and therapy. Journal of the American Medical Association2001285785795. (doi:10.1001/jama.285.6.785).


PanicoALupoliGAFondericoFMarcielloFMartinelliAAssanteRLupoliG. Osteoporosis and thyrotropin-suppressive therapy: reduced effectiveness of alendronate. Thyroid200919437442. (doi:10.1089/thy.2008.0428).


GiustiABaroneAPioliGGirasoleGSiccardiVPalummeriEBianchiG. Alendronate and indapamide alone or in combination in the management of hypercalciuria associated with osteoporosis: a randomized controlled trial of two drugs and three treatments. Nephrology, Dialysis, Transplantation20092414721477. (doi:10.1093/ndt/gfn690).


LindahlKLangdahlBLjunggrenOKindmarkA. Treatment of osteogenesis imperfecta in adults. European Journal of Endocrinology2014171R79R90. (doi:10.1530/EJE-14-0017).


BarbourKELuiLYEnsrudKEHillierTALeBlancESIngSWHochbergMCCauleyJAStudy of Osteoporotic Fractures Research G. Inflammatory markers and risk of hip fracture in older white women: the study of osteoporotic fractures. Journal of Bone and Mineral Research20142920572064. (doi:10.1002/jbmr.2245).


KanisJAJohanssonHOdenAMcCloskeyEV. Guidance for the adjustment of FRAX according to the dose of glucocorticoids. Osteoporosis International201122809816. (doi:10.1007/s00198-010-1524-7).


van StaaTPLeufkensHGCooperC. The epidemiology of corticosteroid-induced osteoporosis: a meta-analysis. Osteoporosis International200213777787. (doi:10.1007/s001980200108).


van StaaTPLeufkensHGAbenhaimLZhangBCooperC. Use of oral corticosteroids and risk of fractures. Journal of Bone and Mineral Research2000159931000. (doi:10.1359/jbmr.2000.15.6.993).


SaagKGEmkeyRSchnitzerTJBrownJPHawkinsFGoemaereSThamsborgGLibermanUADelmasPDMaliceMP. Alendronate for the prevention and treatment of glucocorticoid-induced osteoporosis. Glucocorticoid-Induced Osteoporosis Intervention Study Group. New England Journal of Medicine1998339292299. (doi:10.1056/NEJM199807303390502).


De VriesFBrackeMLeufkensHGLammersJWCooperCVan StaaTP. Fracture risk with intermittent high-dose oral glucocorticoid therapy. Arthritis and Rheumatism200756208214. (doi:10.1002/art.22294).


IsraelEBanerjeeTRFitzmauriceGMKotlovTVLaHiveKLeBoffMS. Effects of inhaled glucocorticoids on bone density in premenopausal women. New England Journal of Medicine2001345941947. (doi:10.1056/NEJMoa002304).


CanalisEMazziottiGGiustinaABilezikianJP. Glucocorticoid-induced osteoporosis: pathophysiology and therapy. Osteoporosis International20071813191328. (doi:10.1007/s00198-007-0394-0).


HofbauerLCGoriFRiggsBLLaceyDLDunstanCRSpelsbergTCKhoslaS. Stimulation of osteoprotegerin ligand and inhibition of osteoprotegerin production by glucocorticoids in human osteoblastic lineage cells: potential paracrine mechanisms of glucocorticoid-induced osteoporosis. Endocrinology199914043824389. (doi:10.1210/endo.140.10.7034).


KimHJZhaoHKitauraHBhattacharyyaSBrewerJAMugliaLJRossFPTeitelbaumSL. Glucocorticoids suppress bone formation via the osteoclast. Journal of Clinical Investigation200611621522160. (doi:10.1172/JCI28084).


PereiraRCDelanyAMCanalisE. Effects of cortisol and bone morphogenetic protein-2 on stromal cell differentiation: correlation with CCAAT-enhancer binding protein expression. Bone200230685691. (doi:10.1016/S8756-3282(02)00687-7).


CanalisE. Wnt signalling in osteoporosis: mechanisms and novel therapeutic approaches. Nature Reviews. Endocrinology20139575583. (doi:10.1038/nrendo.2013.154).


SmithEFrenkelB. Glucocorticoids inhibit the transcriptional activity of LEF/TCF in differentiating osteoblasts in a glycogen synthase kinase-3β-dependent and -independent manner. Journal of Biological Chemistry200528023882394. (doi:10.1074/jbc.M406294200).


DelanyAMDurantDCanalisE. Glucocorticoid suppression of IGFI transcription in osteoblasts. Molecular Endocrinology20011517811789. (doi:10.1210/mend.15.10.0704).


GiustinaAMazziottiGCanalisE. Growth hormone, insulin-like growth factors, and the skeleton. Endocrine Reviews200829535559. (doi:10.1210/er.2007-0036).


DallasSLPrideauxMBonewaldLF. The osteocyte: an endocrine cell … and more. Endocrine Reviews201334658690. (doi:10.1210/er.2012-1026).


GiustinaAVeldhuisJD. Pathophysiology of the neuroregulation of growth hormone secretion in experimental animals and the human. Endocrine Reviews199819717797. (doi:10.1210/edrv.19.6.0353).


CanalisE. Clinical review 83: Mechanisms of glucocorticoid action in bone: implications to glucocorticoid-induced osteoporosis. Journal of Clinical Endocrinology and Metabolism19968134413447. (doi:10.1210/jcem.81.10.8855781).


DiederichSQuinklerMBurkhardtPGrossmannCBahrVOelkersW. 11Beta-hydroxysteroid-dehydrogenase isoforms: tissue distribution and implications for clinical medicine. European Journal of Clinical Investigation200030 (Suppl 3) 2127. (doi:10.1046/j.1365-2362.2000.0300s3021.x).


TanninGMAgarwalAKMonderCNewMIWhitePC. The human gene for 11β-hydroxysteroid dehydrogenase. Structure, tissue distribution, and chromosomal localization. Journal of Biological Chemistry19912661665316658.


CooperMSRabbittEHGoddardPEBartlettWAHewisonMStewartPM. Osteoblastic 11β-hydroxysteroid dehydrogenase type 1 activity increases with age and glucocorticoid exposure. Journal of Bone and Mineral Research200217979986. (doi:10.1359/jbmr.2002.17.6.979).


CooperMSWalkerEABlandRFraserWDHewisonMStewartPM. Expression and functional consequences of 11β-hydroxysteroid dehydrogenase activity in human bone. Bone200027375381. (doi:10.1016/S8756-3282(00)00344-6).


CanalisE. Mechanisms of glucocorticoid-induced osteoporosis. Current Opinion in Rheumatology200315454457. (doi:10.1097/00002281-200307000-00013).


AngeliAGuglielmiGDovioACapelliGde FeoDGianniniSGiorginoRMoroLGiustinaA. High prevalence of asymptomatic vertebral fractures in post-menopausal women receiving chronic glucocorticoid therapy: a cross-sectional outpatient study. Bone200639253259. (doi:10.1016/j.bone.2006.02.005).


GrossmanJMGordonRRanganathVKDealCCaplanLChenWCurtisJRFurstDEMcMahonMPatkarNM. American College of Rheumatology 2010 recommendations for the prevention and treatment of glucocorticoid-induced osteoporosis. Arthritis Care & Research20106215151526. (doi:10.1002/acr.20295).


van StaaTPLaanRFBartonIPCohenSReidDMCooperC. Bone density threshold and other predictors of vertebral fracture in patients receiving oral glucocorticoid therapy. Arthritis and Rheumatism20034832243229. (doi:10.1002/art.11283).


LekamwasamSAdachiJDAgnusdeiDBilezikianJBoonenSBorgstromFCooperCDiez PerezAEastellRHofbauerLC. A framework for the development of guidelines for the management of glucocorticoid-induced osteoporosis. Osteoporosis International20122322572276. (doi:10.1007/s00198-012-1958-1).


DovioAPerazzoloLOsellaGVenturaMTermineAMilanoEBertolottoAAngeliA. Immediate fall of bone formation and transient increase of bone resorption in the course of high-dose, short-term glucocorticoid therapy in young patients with multiple sclerosis. Journal of Clinical Endocrinology and Metabolism20048949234928. (doi:10.1210/jc.2004-0164).


SzappanosATokeJLippaiDPatocsAIgazPSzucsNFutoLGlazERaczKTothM. Bone turnover in patients with endogenous Cushing's syndrome before and after successful treatment. Osteoporosis International201021637645. (doi:10.1007/s00198-009-0978-y).


BuckleyLMLeibESCartularoKSVacekPMCooperSM. Calcium and vitamin D3 supplementation prevents bone loss in the spine secondary to low-dose corticosteroids in patients with rheumatoid arthritis. A randomized, double-blind, placebo-controlled trial. Annals of Internal Medicine1996125961968. (doi:10.7326/0003-4819-125-12-199612150-00004).


GodschalkMLevyJRDownsRWJr. Glucocorticoids decrease vitamin D receptor number and gene expression in human osteosarcoma cells. Journal of Bone and Mineral Research199272127. (doi:10.1002/jbmr.5650070105).


HidalgoAADeebKKPikeJWJohnsonCSTrumpDL. Dexamethasone enhances 1α,25-dihydroxyvitamin D3 effects by increasing vitamin D receptor transcription. Journal of Biological Chemistry20112863622836237. (doi:10.1074/jbc.M111.244061).


AdachiJDBensenWGBianchiFCividinoAPillersdorfSSebaldtRJTugwellPGordonMSteeleMWebberC. Vitamin D and calcium in the prevention of corticosteroid induced osteoporosis: a 3 year followup. Journal of Rheumatology1996239951000.


ReidDMDevogelaerJPSaagKRouxCLauCSReginsterJYPapanastasiouPFerreiraAHartlFFasholaT. Zoledronic acid and risedronate in the prevention and treatment of glucocorticoid-induced osteoporosis (HORIZON): a multicentre, double-blind, double-dummy, randomised controlled trial. Lancet200937312531263. (doi:10.1016/S0140-6736(09)60250-6).


ThomasTHorlaitSRingeJDAbelsonAGoldDTAtlanPLangeJL. Oral bisphosphonates reduce the risk of clinical fractures in glucocorticoid-induced osteoporosis in clinical practice. Osteoporosis International201324263269. (doi:10.1007/s00198-012-2060-4).


SaagKGShaneEBoonenSMarinFDonleyDWTaylorKADalskyGPMarcusR. Teriparatide or alendronate in glucocorticoid-induced osteoporosis. New England Journal of Medicine200735720282039. (doi:10.1056/NEJMoa071408).


BassettJHWilliamsGR. Critical role of the hypothalamic–pituitary–thyroid axis in bone. Bone200843418426. (doi:10.1016/j.bone.2008.05.007).


BassettJHWilliamsGR. The molecular actions of thyroid hormone in bone. Trends in Endocrinology and Metabolism200314356364. (doi:10.1016/S1043-2760(03)00144-9).


BauerDCEttingerBNevittMCStoneKLStudy of Osteoporotic Fractures Research G. Risk for fracture in women with low serum levels of thyroid-stimulating hormone. Annals of Internal Medicine2001134561568. (doi:10.7326/0003-4819-134-7-200104030-00009).


AbrahamsenBJorgensenHLLaulundASNyboMBrixTHHegedusL. Low serum thyrotropin level and duration of suppression as a predictor of major osteoporotic fractures – the OPENTHYRO register cohort. Journal of Bone and Mineral Research20142920402050. (doi:10.1002/jbmr.2244).


CummingsSRNevittMCBrownerWSStoneKFoxKMEnsrudKECauleyJBlackDVogtTM. Risk factors for hip fracture in white women. Study of Osteoporotic Fractures Research Group. New England Journal of Medicine1995332767773. (doi:10.1056/NEJM199503233321202).


VestergaardPMosekildeL. Fractures in patients with hyperthyroidism and hypothyroidism: a nationwide follow-up study in 16,249 patients. Thyroid200212411419. (doi:10.1089/105072502760043503).


AbeEMariansRCYuWWuXBAndoTLiYIqbalJEldeiryLRajendrenGBlairHC. TSH is a negative regulator of skeletal remodeling. Cell2003115151162. (doi:10.1016/S0092-8674(03)00771-2).


van RijnLEPopVJWilliamsGR. Low bone mineral density is related to high physiological levels of free thyroxine in peri-menopausal women. European Journal of Endocrinology2014170461468. (doi:10.1530/EJE-13-0769).


BiondiBWartofskyL. Treatment with thyroid hormone. Endocrine Reviews201435433512. (doi:10.1210/er.2013-1083).


HeemstraKAHamdyNARomijnJASmitJW. The effects of thyrotropin-suppressive therapy on bone metabolism in patients with well-differentiated thyroid carcinoma. Thyroid200616583591. (doi:10.1089/thy.2006.16.583).


SheppardMCHolderRFranklynJA. Levothyroxine treatment and occurrence of fracture of the hip. Archives of Internal Medicine2002162338343. (doi:10.1001/archinte.162.3.338).


TuckSPFrancisRM. Testosterone, bone and osteoporosis. Frontiers of Hormone Research200937123132. (doi:10.1159/000176049).


KrupskiTLSmithMRLeeWCPashosCLBrandmanJWangQBottemanMLitwinMS. Natural history of bone complications in men with prostate carcinoma initiating androgen deprivation therapy. Cancer2004101541549. (doi:10.1002/cncr.20388).


MalcolmJBDerweeshIHKincadeMCDiBlasioCJLamarKDWakeRWPattersonAL. Osteoporosis and fractures after androgen deprivation initiation for prostate cancer. Canadian Journal of Urology20071435513559.


LederBZLeBlancKMSchoenfeldDAEastellRFinkelsteinJS. Differential effects of androgens and estrogens on bone turnover in normal men. Journal of Clinical Endocrinology and Metabolism200388204210. (doi:10.1210/jc.2002-021036).


HofbauerLCKhoslaS. Androgen effects on bone metabolism: recent progress and controversies. European Journal of Endocrinology1999140271286. (doi:10.1530/eje.0.1400271).


PedersonLKremerMJuddJPascoeDSpelsbergTCRiggsBLOurslerMJ. Androgens regulate bone resorption activity of isolated osteoclasts in vitro. PNAS199996505510. (doi:10.1073/pnas.96.2.505).


KasperkCHWakleyGKHierlTZieglerR. Gonadal and adrenal androgens are potent regulators of human bone cell metabolism in vitro. Journal of Bone and Mineral Research199712464471. (doi:10.1359/jbmr.1997.12.3.464).


HuberDMBendixenACPathrosePSrivastavaSDiengerKMShevdeNKPikeJW. Androgens suppress osteoclast formation induced by RANKL and macrophage-colony stimulating factor. Endocrinology200114238003808. (doi:10.1210/endo.142.9.8402).


BehreHMKlieschSLeifkeELinkTMNieschlagE. Long-term effect of testosterone therapy on bone mineral density in hypogonadal men. Journal of Clinical Endocrinology and Metabolism19978223862390. (doi:10.1210/jcem.82.8.4163).


GiustiABianchiG. Treatment of primary osteoporosis in men. Clinical Interventions in Aging201510105115. (doi:10.2147/CIA.S44057).


PepeneCECrisanNComanI. Elevated serum receptor activator of nuclear factor κB ligand and osteoprotegerin levels in late-onset male hypogonadism. Clinical and Investigative Medicine. Médecine Clinique et Experimental201134E232.


ParthanAKruseMAgodoaISilvermanSOrwollE. Denosumab: a cost-effective alternative for older men with osteoporosis from a Swedish payer perspective. Bone201459105113. (doi:10.1016/j.bone.2013.11.002).


PavoneMEBulunSE. Clinical review: The use of aromatase inhibitors for ovulation induction and superovulation. Journal of Clinical Endocrinology and Metabolism20139818381844. (doi:10.1210/jc.2013-1328).


LegroRSBrzyskiRGDiamondMPCoutifarisCSchlaffWDCassonPChristmanGMHuangHYanQAlveroR. Letrozole versus clomiphene for infertility in the polycystic ovary syndrome. New England Journal of Medicine2014371119129. (doi:10.1056/NEJMoa1313517).


ColemanREBanksLMGirgisSIKilburnLSVrdoljakEFoxJCawthornSJPatelASnowdonCFHallE. Skeletal effects of exemestane on bone-mineral density, bone biomarkers, and fracture incidence in postmenopausal women with early breast cancer participating in the Intergroup Exemestane Study (IES): a randomised controlled study. Lancet Oncology20078119127. (doi:10.1016/S1470-2045(07)70003-7).


EastellRHannonRACuzickJDowsettMClackGAdamsJEATAC Trialists' group. Effect of an aromatase inhibitor on BMD and bone turnover markers: 2-year results of the Anastrozole, Tamoxifen, Alone or in Combination (ATAC) trial (18233230). Journal of Bone and Mineral Research20062112151223. (doi:10.1359/jbmr.060508).


HowellACuzickJBaumMBuzdarADowsettMForbesJFHoctin-BoesGHoughtonJLockerGYTobiasJS. Results of the ATAC (Arimidex, Tamoxifen, Alone or in Combination) trial after completion of 5 years' adjuvant treatment for breast cancer. Lancet20053656062. (doi:10.1016/S0140-6736(05)74803-0).


KaunitzAMAriasRMcClungM. Bone density recovery after depot medroxyprogesterone acetate injectable contraception use. Contraception2008776776. (doi:10.1016/j.contraception.2007.10.005).


CannCE. Bone densitometry as an adjunct to GnRH agonist therapy. Journal of Reproductive Medicine199843321330.</