Heart Failure as a Risk Factor for Osteoporosis and Fractures
Although heart failure (HF) and osteoporosis are common diseases, particularly in elderly populations, patients with HF have an increased risk for osteoporosis. The relationship of HF with osteoporosis is modified by gender and the severity of HF. In addition, shared risk factors, medication use, and common pathogenic mechanisms affect both HF and osteoporosis. Shared risk factors for these 2 conditions include advanced age, hypovitaminosis D, renal disease, and diabetes mellitus. Medications used to treat HF, including spironolactone, thiazide diuretics, nitric oxide donors, and aspirin, may protect against osteoporosis. In contrast, loop diuretics may make osteoporosis worse. HF and osteoporosis appear to share common pathogenic mechanisms, including activation of the renin-angiotensin-aldosterone system, increased parathyroid hormone levels, and/or oxidative/nitrosative stress. HF is a major risk factor for mortality following fractures. Thus, in HF patients, it is important to carefully assess osteoporosis and take measures to reduce the risk of osteoporotic fractures.
KeywordsHeart failureOsteoporosisFracturesHypovitaminosis DRenal diseaseDiabetesSpironolactoneThiazide diureticsNitric oxide donorsAspirinLoop diureticsRenin-angiotensin-aldosterone systemParathyroid hormone levelsOxidative/nitrosative stress
Congestive heart failure (HF) and osteoporosis are prevalent diseases, particularly in an aging population [1, 2]. Worldwide, 20 million persons have HF  and more than 75 million have osteoporosis . The incidence of acute hospitalizations for both major osteoporotic fractures and major cardiovascular events, including heart failure, increased in both men and women from 2000 to 2008 . Approximately 670,000 incident diagnoses of HF occur in the United States (U.S.) every year . In 2005, 2,000,000 new osteoporosis-related fractures occurred in the U.S. . Advanced age is a critical risk factor for both HF and hip fractures . In a study of over 200,000 Medicare beneficiaries with heart failure stratified into 3 age groups (66–75;76–85 and 86+ years), hip fractures were common and were more prevalent among the oldest relative to the youngest group (9.5 % vs 1.9 %) .
Heart Failure, Bone Mineral Density, and Body Composition
Bone mineral density (BMD) is a powerful predictor of fracture risk. For every standard deviation decrease in BMD by dual energy X-ray absorptiometry (DXA), fracture risk approximately doubles [9–11]. Body composition can also be estimated by DXA, and body composition has recently been associated with fracture risk [12, 13].
There is increasing evidence that HF is a risk factor for osteoporosis [14•, 15, 16•, 17•]. Experimental studies in mice deficient in Apo lipoprotein E demonstrate that arterial calcifications and aortic valve calcifications correlate directly with BMD losses . In humans, the relationship of HF to BMD has been addressed in a number of studies [16•, 19•, 20–22]. The weight of evidence supports the hypothesis that patients with HF are at risk for low BMD [14•, 16•, 20–23]. A significant inverse relationship between HF and BMD at the total hip, femoral neck, total body, and lumbar spine has been reported [19•, 24]. The few studies that have examined longitudinal changes in BMD suggest that there is accelerated loss of BMD in patients with HF compared with controls [22, 25]. In 1 report, significant bone loss over time (≥1 %/year decline in bone mineral content) occurred in 35 % of men with HF . Another study that compared rates of BMD loss in patients with and without HF determined that there were accelerated rates of femoral neck BMD loss in those patients with HF and renal insufficiency compared with patients without HF . Body composition is a risk factor for the development of HF, and recently, higher levels of visceral adiposity have been associated with the development of HF . Visceral adiposity estimated by DXA may also be associated with osteoporosis risk, with higher levels of visceral adiposity associated with lower BMD .
Heart Failure and Osteoporotic Fractures
The weight of evidence also suggests that there is a positive relationship between osteoporotic fractures, in particular, hip fractures, and HF. In the Cardiovascular Health Study (CHS), a trend for a relationship between incident HF and incident hip fractures was reported . Similarly, in the Rochester Epidemiology Project, both prevalent and incident osteoporotic fractures were more common in heart failure patients compared with age and sex matched community controls, and this increased risk was almost entirely attributable to hip fractures [14•]. In a study of over 31,000 twins followed through the National Patient Registry in Sweden, twins with heart failure had a multivariable-adjusted hazard ratio for hip fractures of 3.74 . Osteoporotic fracture sites other than the hip have also been associated with HF [16•, 17•]. In 1 study, approximately one-tenth of patients with HF were found to have vertebral compression fractures on radiographs, one half of these had multiple fractures, and the majority (85 %) was not on treatment for osteoporosis [17•]. In a population based study in Manitoba, Canada including over 45,500 adults with 2,703 incident osteoporotic fractures, which linked BMD data from 1998–2009 in those age 50+ years with administrative databases, there was a 30 % increase in major fractures in persons with HF [16•]. Importantly, in this study, the increased risk for osteoporotic fractures was not only independent of osteoporosis risk factors, comorbidities, and medications, but was also independent of BMD measurements of the total hip [16•].
Covariates for the Relationship of Heart Failure and Osteoporosis
The association of HF with osteoporosis may be dependent on a number of factors including gender, the severity of HF, and the type of HF (systolic vs diastolic). Some studies have noted that the relationship between HF and low BMD is only present in males [22, 24], although others suggest that this relationship exists regardless of gender [16•, 21, 30]. More severe HF has been correlated with lower BMD measurements in a number of studies [19•, 21, 31, 32•]. In contrast, 1 report suggested that there was no association between the severity of heart failure and incident major fractures, although in this study, the severity of heart failure was determined on the basis of hospitalizations [16•]. The relationship of type of HF to BMD is controversial, with 1 study suggesting that patients with systolic HF have a slightly higher total BMD , while another study reported that there was no relationship between type of HF and incident hip fractures . Among patients with systolic heart failure, lower ejection fractions have been associated with lower BMD measurements [19•, 21, 31, 32•].
Shared Risk Factors for Heart Failure and Osteoporosis
The relationship between HF, BMD, and fractures
BMD and HF
Fracture and HF
Decreased BMD total hip [19•]
Increased risk hip fracture [16•]
Decreased BMD femoral neck [19•]
>30 % Increased risk of any major fracture (humerus, vertebral, hip, and forearm) [16•]
Decreased BMD total lumbar spine [19•]
Increased risk vertebral compression fx [16•]
Severity of HF:
Severity of HF:
Decreased BMD of femoral neck and total body with increase in NYHA class [32•]
No relationship between severity (measured by hospitalizations) and major fractures (humerus, vertebral, hip, and forearm) [16•]
Decreased BMD of femoral neck and total body with worsening ejection fraction [32•]
There are a number of shared risk factors for HF and osteoporosis [26, 35–54]. Older age [35–37], and the presence of renal disease [26, 38, 39] and/or diabetes mellitus [40, 41] are shared risk factors for both conditions. Vitamin D receptors are present on both bone  and cardiac muscle , and vitamin D deficiency, an established risk factor for osteoporosis [44, 57], has also been reported to contribute to HF . However, The Institute of Medicine (IOM) found that current evidence supported a role for vitamin D in bone health but not in other health conditions. Likewise, the Endocrine Society Task Force, in their recent review of the evaluation, treatment, and prevention of vitamin D deficiency, concluded that there was not sufficient evidence that vitamin D should be used for cardiovascular protection [59•]. In the Framingham Heart Study, obesity increased the risk for HF by 5 % for men and 7 % for women for each increment of 1 in body-mass index . The relationship of obesity to osteoporosis, however, is controversial. In 1 study, lower BMD was reported in obese compared with non-obese subjects . However, most studies have reported that BMD is higher in obese compared with non-obese subjects [51, 60]. The relationship of BMI to osteoporotic fractures is complex, and is likely site dependent, with obesity decreasing the risk for hip and pelvic fractures and increasing the risk for humeral fractures and ankle fractures [51–53, 60, 61]. This increased risk for humeral fractures in obesity is important, because there is a high incidence rate for humeral fractures among elderly patients in the U.S. . Moreover, humeral fractures are associated with substantial morbidity . Among women, postmenopausal status is a risk factor for both HF and osteoporosis [42, 43]. In elderly men, in particular, men with HF, hypogonadism is a prevalent condition [64–68]. Hypogonadism is a risk factor for osteoporosis, and testosterone replacement can increase BMD [47, 48]. In addition, symptoms of heart failure can be improved by testosterone administration in hypogonadal patients . In a recent meta-analysis including 4 randomized, double-blind, placebo controlled trials involving 198 subjects with HF, NYHA class, and exercise capacity as assessed by the 6-minute walk test, incremental shuttle walk test, and peak maximum oxygen consumption, were significantly improved in those patients treated with testosterone compared with placebo, without an increase in adverse cardiovascular events [69•]. In contrast, in another placebo-controlled, randomized trial of elderly men with hypogonadism, testosterone supplementation increased adverse cardiovascular events, including HF .
HF Medication Use and Risk of Osteoporosis
The relationship of cardiac medications to osteoporosis
Effect on osteoporosis risk
Mechanism of action
↓ Renal Ca excretion
↓ Renal Ca excretion, ↑ K stores, ↓ aldosterone-mediated bone loss
NO stimulation of osteoblastic activation
Inhibition of osteoclastogenesis and stimulation of osteoblastic activity
↑ Renal Ca excretion
PPAR-γ activated osteoclast differentiation
Angiotensin receptor blockers
Possible action at the angiotensin receptor
Angiotensin action on osteoblasts or calcium metabolism
↓ Mevalonate, which results in enhanced osteoblastic differentiation
β-Adrenergic receptor inhibition
Mechanisms of Action of HF Medications on Osteoporosis
There are a number of mechanisms by which medications used to treat HF may impact on osteoporosis risk. One such mechanism is through changes in calcium excretion, since higher calcium excretion is associated with lower BMD . Thiazide diuretics’ inverse association with osteoporosis may be mediated at least in part through decreases in calcium excretion [71•, 98]. It follows that increases in calcium excretion by loop diuretics may negatively affect osteoporosis . Spironolactone has a number of mechanisms by which it might be beneficial for osteoporosis. In addition to decreases in calcium excretion, spironolactone increases potassium stores , and potassium intake is positively related to BMD [100, 101]. In animal models, aldosteronism is associated with tibia and femoral bone loss over time and reduction in bone strength to flexor stress . Thus, it is possible that inhibition of the aldosterone receptor by spironolactone may reverse/prevent bone loss. The β-adrenergic receptor signaling pathway has been thought to play a role in bone regulation since chemical sympathectomy in rats impairs bone resorption by reducing the bone resorption surface and impairing osteoclast access to the bone surface . Osteoblasts possess β-2-adrenergic receptors [103, 104]. In animal studies, β-adrenergic activation in osteoblasts by norepinephrine or isoproterenol leads to increase in cAMP in a similar fashion as that stimulated by parathyroid hormone . Epinephrine has been shown to increase receptor activator of nuclear factor-kappa B ligand (RANKL) expression, a dominant regulator of bone resorption, by activation of β-adrenergic receptors [106, 107]. Thus, β-adrenergic receptor activation in this setting activates osteoclastogenesis. The mechanism of action of nitric oxide (NO) donors on the skeleton is thought to be mediated by increases in osteoblastic bone formation. NO stimulates guanylate cyclase which increases cGMP, and activates protein kinase G (PKG) . PKG activate the kinase systems leading to extracellular regulated kinase induction of c-fos, fra-1, fra-2, and fosB/ΔfosB, which play a key role in the osteoblast anabolic activity . The mechanism by which aspirin use is positively associated with BMD is not clearly understood. One of the major effects of aspirin is to inhibit prostaglandin (PG) production by acetylating a critical serine in the arachidonic acid binding site of cyclooxygenase (COX) 1 and 2 [109, 110]. The in vitro effects of aspirin to inhibit osteoclast formation are similar to those seen with deficiency or inhibition of COX-2 , while the in vitro effects of aspirin to stimulate osteoblastic mineralization are the opposite of the effects of COX-2 deficiency or inhibition . Thiazolidinediones are thought to exert their antidiabetic effect by activating peroxisome proliferator-activated receptor-γ nuclear receptor (PPAR-γ), which increases glucose and lipid uptake, glucose oxidation, and decreases free fatty acid concentrations thereby decreasing insulin resistance . PPAR-γ may promote osteoclast recruitment and differentiation from hematopoietic progenitors by controlling expression of c-fos [113, 114]. Statins inhibit HMG-CoA reductase and reduce intracellular mevalonate. In an animal model of osteoporosis, statins were shown to reduce oxidative stress and to restore NO formation .
Osteoporosis Medication Use and Risk of HF
The relationship of pharmacological treatments for osteoporosis to cardiovascular disease (CVD) has been investigated in a number of studies [26, 45, 77, 78]. Adequate calcium intake is recommended in all patients with osteoporosis . However, there has been substantial controversy concerning the role of calcium supplementation in CVD. Increased rates of myocardial infarction in women supplemented with calcium compared with placebo group have been reported [77, 78]. However, more recent studies suggest that calcium supplementation does not alter the presence or amount of coronary calcifications , a known risk factor for CVD [117–120]. There are no studies that link calcium supplementation specifically to HF, but hypocalcemia has been associated with a reversible dilated cardiomyopathy .
A number of pharmacological agents [including selective estrogen receptor modulators (SERMs), estrogen/estrogen with progesterone, calcitonin, bisphosphonates, PTH, and RANKL inhibitors] are used to treat osteoporosis. Raloxifene, a SERM used to treat postmenopausal osteoporosis, has not been linked to any risk of cardiovascular disease . However, there are no studies on the effects of raloxifene specifically in HF. Endogenous estrogen is noted to be protective against myocardial infarction (MI) in premenopausal women [123, 124]. On the other hand, pharmacological supplementation of postmenopausal women with estrogen plus progestin has been associated with increased coronary risk , and elevated estrogen levels have been shown to worsen arteriosclerosis by increasing intimal thickness . The results of estrogen therapy on cardiovascular outcomes may depend on the timing of administration of estrogen therapy in relationship to menopause with potential benefits in early administration in the menopausal period and potential adverse events if started late in menopause, (i.e., the so called “timing hypothesis”) . Calcitonin is also used to treat osteoporosis. There are no reports that administration of calcitonin for osteoporosis is associated with HF events. Early reports suggested that there might be a link between bisphosphonate use and risk of atrial fibrillation [128–130]; however; more recent studies have found no increased risk of atrial fibrillation in patients treated with bisphosphonates [131–137]. In accord with this, in 2008, The Food and Drug Administration (FDA) concluded that there was no clear association between bisphosphonate exposure and the rate of serious or non-serious atrial fibrillation events . Recently, of interest, in 1 study, atrial fibrillation itself was found to be a risk factor for vertebral fractures in patients with HF [17•]. There are no reports linking bisphosphonate use to HF; however, in a recent retrospective cohort study, use of alendronate was associated with increased risk of atherosclerosis and acute MI . Teriparatide, the 1-34 amino end of PTH, inhibits vascular calcification and exerts beneficial actions at early stages of macro vascular disease responses to diabetes and dyslipidemia in animal studies . Still, whether it affects vascular disease in humans is unknown. The extra-skeletal effects of Denosumab, the most recent agent available to treat osteoporosis, have yet to be assessed .
Common Pathogenic Mechanisms for Heart Failure and Osteoporosis
Beyond common risk factors, HF and osteoporosis may share a common underlying pathology. In support of this, in a large cohort study, patients with HF followed for 4 years had a 30 % increased risk of having a fracture compared with patients without HF, and this risk was present after adjustment for shared risk factors and use of medications for treatment of heart failure [16•]. There are a number of possible common pathways for HF and osteoporosis. In the authors’ opinion, the most likely pathways for a shared pathogenesis between these 2 disease states would be mediated through 1 or more of the following: Parathyroid hormone (PTH), oxidative stress, and/or the renin-angiotensin system. Increased levels of PTH may play a central role in both HF and osteoporosis [19•, 23, 24, 142, 143•, 144]. In patients with HF, increased PTH levels were correlated with reduced total body and femoral neck BMD and, levels of PTH were also associated with HF severity as assessed by cardiac index, vo2 peak, and wedge pressure [19•]. In HF, increased parathyroid hormone accounts for paradoxical intracellular calcium overloading in a number of tissues and consequent systemic induction of oxidative stress . Oxidative stress is at least partly responsible for the development and progression of heart failure . Reactive oxygen species directly impair contractile function by modifying proteins central to excitation-contraction coupling, activate hypertrophy signaling kinases and transcription factors and mediate apoptosis, stimulate fibroblast proliferation, and activate MMP . Oxidative stress also plays a key role in osteoporosis. In in vitro and animal studies, superoxide deficiency has been associated with low turnover osteoporosis . An imbalance between oxidant and antioxidant status is associated with increased osteoclastic and decreased osteoblastic activity . In 1 study, patients with osteoporosis had lower glutathione reductase enzyme activity, higher malondialdehyde (MDA), and higher nitric oxide (NO) than patients without osteoporosis . In this study, there was no significant relationship between NO levels and lumbar spine or femoral neck BMD although MDA levels were negatively associated with femoral neck BMD . Genetic associations of reduced femoral neck BMD with Glutathione S-transferases mu3 (GSTM3) alleles, which are a group of detoxifying enzymes that eliminate oxidative stress-related products, have recently been identified [151, 152]. The utility of using a micronutrient, zinc, to protect against oxidative stress and to potentially impact on osteoporosis was investigated in a recent study . In this study, administration of zinc inhibited release of cytochrome c from mitochondria to cytosol in zinc + H(2)O(2)-treated cells. The authors suggested that protecting osteoblasts from oxidative stress using zinc supplementation is a potential mechanism to prevent osteoporosis . There is a critical role for the renin-angiotensin-system in both HF  and osteoporosis . The Angiotensin II receptor is a cell surface G-protein coupled receptor which upon activation leads to intracellular increases in cyclic AMP (cAMP) . Activation of cAMP signaling pathway can upregulate receptor activator of RANKL expression in osteoblasts  and subsequently activate osteoclasts . In addition, angiotensin II regulates blood flow in both bone capillary and blood vessels’ endothelium, providing a functional link between osteoporosis and atherosclerosis .
Adiponectin, an adipocytokine secreted by adipocytes, also plays a role in both cardiovascular disease  and osteoporosis . Elevated adiponectin levels in HF reflect an attempt by the body to compensate for the impaired metabolism in this disease . Relative to osteoporosis, adiponectin stimulates osteoblast proliferation and differentiation . A recent meta-analysis including 51 studies found that adiponectin is the most relevant adipokine negatively associated with BMD independent of gender and menopausal status . Furthermore, higher adiponectin levels are associated with increased fracture risk independent of body composition and BMD . In HF, higher adiponectin levels are associated with lower BMD .
Heart Failure, Osteoporosis, and Mortality
Both HF and osteoporosis substantially increase mortality across race and gender [161, 162]. Mortality following HF or hip fractures may be associated with health care expenditures for treatment of these conditions, with mortality lower in hospitals that spend more for these conditions. This is true for systems with universal access to health care  and in the U.S. .
Osteoporosis may have substantial prognostic significance in HF. Among patients with HF, lower BMD was associated with increased rates of death, implantation of a left ventricular assist device, and/or inotrope dependency [19•]. When HF and osteoporosis are both present in a patient, subsequent mortality is more than additive [29, 163]. In men and women with HF, those with both HF and hip fractures have a two-fold increased mortality relative to persons with heart failure without hip fractures . HF is also a major risk factor for mortality following hip [163–165] and distal femur fractures .
Care of all individuals with osteoporotic fractures is critically important . In recognition of the success of the American Heart Association’s U.S. National Program to improve the use of β-blockers following myocardial infarction (MI) to reduce MI events (“Get with the Guidelines”), “Own the Bone”, a multidisciplinary quality improvement program, was launched by the orthopedic community to enhance care of patients age 50 and older with fragility fractures . Further studies are needed to determine whether improved management of HF among hip fracture patients can improve survival .
In summary, HF and osteoporosis are prevalent, serious conditions that share common risk factors, and selected aspects of a common pathogenesis. Patients with HF should be considered at risk for osteoporosis, especially in those with more severe HF. Careful screening for osteoporosis with modification of potential risk factors and appropriate medication usage are important considerations in patients with HF.
No potential conflicts of interest relevant to this article were reported.