Nutrition in Cardioskeletal Health^,,

Abstract

Bone and heart health are linked through a variety of cellular, endocrine, and metabolic mechanisms, including the bidirectional effects of mineral-regulating hormones parathyroid hormone and fibroblast growth factor 23. Nutrition plays an important role in the development of both cardiovascular and bone disease. This review describes current knowledge on the relations between the cardiovascular system and bone and the influence of key nutrients involved in mineral metabolism—calcium, vitamin D, and phosphorus—on heart and bone health, as well as the racial/ethnic differences in cardiovascular disease and osteoporosis and the influence that nutrition has on these disparities.

Introduction

Cardiovascular disease (CVD)⁷ and osteoporosis are common age-related diseases that are influenced by nutrition. In particular, calcium, vitamin D, and phosphorus have become prominent as nutrients that play roles in both heart and bone health. This review gives an overview of current knowledge of the cardioskeletal relation and explores aspects of the influence of calcium, vitamin D, and phosphorus on heart and bone, as well as how race and ethnicity might modify the influence of nutrition in CVD and osteoporosis.

The Cardio-Skeletal Relation

Vascular calcification and arteriosclerosis.

Vascular calcification was once thought of as only a passive process of dead and dying cells but is now being recognized as a highly regulated form of matrix mineralization (1). Deposition and egress of vascular calcium-containing salts (phosphates, phospholipids, and carbonates) are controlled by the integrated actions of circulating mineralization inhibitors, paracrine cues driving osteo/chondrogenic cell potential and microvesicle release, and vascular remodeling processes that impact both matrix and mineral. The relation between vascular calcification and arteriosclerotic disease has been increasingly appreciated since the 1850s when Virchow (2) described the histopathologic picture of atherosclerotic disease observed at autopsy as a lipid-laden, intimally oriented, lumen-deforming inflammatory lesion with bone-like sclerotic changes. Although disagreeing with Virchow on a number of other issues, von Rokitansky also noted bone-like morphologic features in advanced atherosclerotic lesions, as discussed by Mayerl et al. (3). Mönckeberg (4) highlighted a different type of vascular mineralization almost 5 decades later: a calcific sclerosis of the arterial tunica media that did not impinge upon the vessel lumen. This medial artery calcification was noted to be more prevalent in diabetes, in chronic kidney disease (CKD), and with aging (5). However, the clinical connections between vascular calcification and arteriosclerosis—the hardening of the conduit arteries that impairs conduit vessel function even in the absence of atherothrombosis or lumen stenosis (5, 6)—has been appreciated more recently. The experimental evidence that diastolic blood flow was substantially impaired by aortic stiffness was conceptually articulated by Frank in 1899 but was not widely available until the 1960s [reviewed by Westerhof (7)]. This Windkessel physiology—the capacity of elastic conduit vessels to store some kinetic energy as potential energy during systole and then release it as kinetic energy to drive perfusion during diastole (7)—has profound implications. While increasing systolic blood pressure and myocardial workload, vascular stiffening that impairs the arterial Windkessel reduces diastolic coronary and distal tissue perfusion (7). This can be well appreciated in the central nervous system, where white matter hyperintensity lesion volume (8), a marker of ischemic flow and progressive cognitive decline (9), tracks aortic stiffness independent of other risk factors including systolic blood pressure (8). Arterial stiffness/compliance is a composite of material properties (smooth muscle myosin-actin engagement; extracellular matrix accumulation, integrity, and cross-linking; mineral deposition, etc.) and geometric properties (wall thickness, lumen diameter) (5). For a variety of reasons, the interrelations between vascular calcium accrual and arterial stiffness are likely to be bidirectional. Indeed, in preclinical disease models of diabetic arteriosclerosis, genetic manipulations that increase or decrease arterial calcium accrual do increase and decrease, respectively, arterial stiffness (10, 11). However, as first highlighted by Engler et al. (12), the mechanocrine environment experienced by multipotent mesenchymal progenitors capable of osteochondrogenic fates markedly influences cellular differentiation. Stiffer matrices suppress myogenic potential while simultaneously enhancing osteogenic and fibrogenic potential of human multipotent mesenchymal cells. Thus, the propensity of proliferating vascular mesenchymal cells to adopt an osteogenic fate and thereby mineralize may represent a feed-forward mechanism reflecting increasing stiffness in the inflamed arterial microenvironment (13). Chen and Simmons (13) have convincingly demonstrated this principle using valve interstitial cells. This relation may explain why tibial artery calcification scores outperform ankle-brachial indexes in predicting the progression of limb ischemia to severity requiring amputation (14); arterial calcium load not only contributes to arteriosclerotic stiffening but also functions as a “biomarker” reflecting the degree of regional tissue stiffness and impaired conduit vessel Windkessel function (15). It remains to be determined the extent to which the enhancement of arterial calcium egress will restore arterial compliance and normal distal tissue perfusion.

Cellular mechanisms of vascular mineralization.

As mentioned above, advanced atherosclerotic lesions have bone-like tissue characteristics. In more recent studies of valve and vascular calcification, ∼15% of advanced lesions exhibit woven bone formation with marrow elements (16). The cellular contributions to true ectopic ossification in advanced lesions, however, may differ substantially from the early lesions forming during the initiation of vascular mineral deposition. Circulation progenitors capable of ectopically recapitulating the entire hematopoietic niche functions of bone may be required (17, 18). Boström et al. (19) first highlighted the role of regional progenitors, the mural pericyte population, as an oxylipid- and cytokine-activated osteogenic cell capable of initiating arterial mineral deposition in the absence of true ectopic bone formation. As in bone, alkaline phosphatase-positive and annexin-positive mineralizing phospholipid matrix vesicles initiate the extracellular calcium deposition with epitaxial growth (20). Importantly, although alkaline phosphatase/tissue nonspecific alkaline phosphatase (TNAP) reduces phospho-osteopontin and pyrophosphate mineralization inhibitors in the local milieu (21, 22) and vascular TNAP transgenesis is sufficient to drive arterial calcification (23), annexin-containing vesicles can nucleate mineralization without antecedent TNAP activity (24). The vascular smooth muscle cell (VSMC) has multiple important contributions. In response to elevated extracellular ionized calcium and phosphate, the VSMC lineage elaborates the extracellular vesicles capable of initiating mineralization (24). Moreover, viable VSMC can phagocytose these vesicles, thereby restricting the number of mineralizing foci nucleating deposition (20). Hyperphosphatemia-induced VSMC apoptosis substantially curtails this important vascular editing function. Furthermore, during dyslipidemia, an osteochondrogenic VSMC transdifferentiation occurs, responsible for ∼80% of the osteogenic cells accruing in vascular preclinical calcification models (25). Regional paracrine cues provided by adventitial (26) and circulating osteoprogenitors (27) have emerged as part of the vascular reprogramming. With diabetes, induction of the endothelial-mesenchymal transition may not only disrupt endothelial-VSMC interactions that preserve the VSMC phenotype but also generate substantial numbers of osteoprogenitors capable of contributing to vascular calcium load (28). As in bone repair following fracture, the relative contributions of each of these osteoprogenitor pools may differ dependent upon anatomic venue. Indeed, elegant studies from St. Hilaire et al. (29) demonstrate that lower extremity peripheral arterial calcification and VSMC-TNAP activity is actively restrained in humans by genetic programs distinct from those governing aortic and coronary calcification. However, VSMC activity of the master osteochondrogenic transcription factor, Runt related transcription factor 2 (Runx2/Cbfal), appears vitally important to all vascular sclerotic responses observed to date. Other transcription factors [mothers against decapentaplegia homolog 1/5 (Smad1/5), muscle segment homeobox homolog 1/2 (Msx1/2), osterix (Sp7/Osx), nuclear factor of activated T-cells (NFAT), sex determining region of the Y chromosome box 9 homolog (Sox9)] inductively regulated by the bone morphogenetic proteins and Wnt ligands that regionally coordinate matrix mineralization mediate their actions in great part via modulation of Runx2/Cbfa1-driven programs (30–32).

Features of vascular mineralization resemble that of calcifying granuloma, a component of the innate immune system that helps contain pulmonary mycobacterial and fungal infections (33). In a TNF-, oxylipid-, and bone morphogenetic protein-dependent fashion, pulmonary vascular mesenchymal cells, endothelial cells, T-cells, and monocyte/macrophage lineage cells direct a rigid, calcified extracellular matrix that physically restricts disease expansion. Intriguingly, the foreign lipids of these pathogens and oxylipids from LDL cholesterol activate Toll-like receptor signals that may hold promise for therapeutic intervention in arteriosclerotic disease (34). Moreover, as in the lung, cells of the monocyte/macrophage lineage program TNAP production by mineralizing progenitors, and recent data indicate that this lineage, too, can elaborate matrix vesicles to nucleate mineralization (35). Although T-cells are vital to the regulation of skeletal mineralization, their contributions to vascular mineral metabolism remain woefully understudied.

The bone-vascular connection in arteriosclerosis and arterial calcification: Clues from the “perfect storm” of CKD.

Bone never forms without vascular interactions. Indeed, during endochondral ossification, the mineralizing osteoprogenitor tracks along microvasculature that penetrates the initially avascular cartilaginous template (36). The osteoprogenitor niche of bone thereafter continues to reside in the perivascular sinusoidal marrow space (18). As discussed above, the macrovascular adventitia and media also create a “niche-like” venue for osteoprogenitors. Moreover, the vasculature provides the route for influx and egress of calcium and phosphate in response to metabolic, morphogenetic, and mechanical needs. Thus, the notion that active biomineralization is regulated in orthotopic and heterotopic venues via vascular cues is not surprising. However, the “perfect storm” of perturbed calcium phosphate homeostasis CKD and its accompanying mineral and bone disorder (CKD-MBD) has highlighted an emerging role for the skeleton in vascular health and homeostasis (37). The earliest changes in calcium phosphate metabolism noted with declining renal function encompass elevations in circulating fibroblast growth factor (FGF) 23, a bone-derived phosphaturic hormone that reduces parathyroid hormone (PTH) secretion and restricts renal calcitriol production (38). Actions in the kidney and the parathyroid require FGF receptor 1 and the coreceptor, Klotho. Loss of FGF23/Klotho signaling results in massive phosphate- and calcitriol-driven arterial calcification (39). High levels of endogenous, renally derived calcitriol, as occurs with FGF23 signaling-deficient mice (39), has direct procalcific actions in VSMC normally held in check by the PTH/PTH-related peptide receptor (PTH1R) (40). In the short term, the upregulation of FGF23 signaling with the phosphate retention of early CKD is a highly desirable homeostatic response, because 1) serum phosphate at elevated levels is a vascular toxin (41) that conveys substantial cardiovascular risk at any level of renal function (42), and 2) as long as glomerular filtration and the capacity for urine generation is preserved, net excretion of phosphate can be achieved to protect the vasculature and other tissues. However, FGF23 also induces myocardial hypertrophy of the left ventricle, independent of Klotho coreceptor actions (43). Because left ventricular hypertrophy predisposes to heart failure and sudden death in CKD, the long-term consequences of elevated bone-derived FGF23 become maladaptive, a state wherein abnormal bone endocrine signals elicit CVD (44). The extent to which this bone-vascular axis contributes to CVD in the absence of overt renal failure remains to be established. However, again, at any level of renal function, increased fasting phosphate levels portend increased CVD.

London et al. (45–47) have identified additional relations between bone homeostasis and vascular calcium load in CKD. In their early studies in humans, the presence of either atherosclerotic intimal calcification or medial artery calcification was shown to predict early mortality in dialysis patients as compared with the longevity enjoyed by the fortunate minority without vascular mineralization (45). Using dynamic histomorphometry, they showed that those individuals with low turnover bone disease exhibit the greatest arterial calcium load (47). Further stratification of vascular disease by ankle-brachial indexes revealed that the presence of peripheral arterial disease tracked impaired skeletal osteoblast synthetic function and apparent resistance to ambient PTH tone (46). Intermittent administration of the bone anabolic agonist PTH(1–34) can reduce arterial calcium deposition in preclinical models (48, 49), suggesting that maintaining normal skeletal osteoblast synthetic function is important for vascular health. Mathew et al. (50) made similar observations with another bone anabolic strategy. However, the arterial vasculature expresses abundant PTH1R, which is bioactive and directly regulates VSMC proliferation (51) and matrix metabolism (48). In the vasculature, paracrine PTH-related peptide vasodilatory signaling in response to mechanical stretch is likely most physiologically relevant (52). The extent to which skeletal compared with VSMC PTH1R signaling contributes to preservation of vascular health is an area of ongoing investigation.

Calcium phosphate nutritional regulation of bone-vascular interactions: Challenges, opportunities, and future directions.

As a consequence of dietary preferences, supplements, medications, and food additives, the intakes of calcium and phosphorus can vary dramatically in the population (53). Because of the direct effects of calcium and phosphate on VSMC mineral metabolism and indirect effects via calciotropic hormones with potent vasculotropic actions, a better understanding of the impact of nutrition and genomics is clearly needed. Declining renal function clearly perturbs the bone-vascular axis in great part via phosphate metabolism, and phosphate binders play an important role in controlling absorption in our phosphorus-laden diets (54). However, a didactic example of the problems faced in CKD-MBD management becomes apparent when calcium-based phosphate binders are implemented; as compared with a calcium-independent phosphate binder, calcium acetate substantially reduces PTH levels, concomitantly reduces bone mass, and increases vascular calcium load in patients on renal replacement therapy (55). The extent to which the declining renal function of aging or diabetes impacts dietary calcium and phosphorus recommendations and vascular health remains to be established (42). The optimal “set points” for PTH/PTH1R and FGF23/FGF receptor 1 signaling tone with respect to vascular health have yet to be established. As with most endocrine physiology, a biphasic relation appears to exist wherein excessive or deficient signaling in response to metabolic need creates challenges to health. In kl/kl mice (Klotho-deficient), a vascular senescence model driven in part by hyperphosphatemia in the setting of normal renal function, the premature vascular calcification tracks induction of Runx2/Cbfa1 and NFAT5 (56), where the latter is an osmoregulated member of the osteochondrogenic NFAT family. Ammonia accumulates in the cytosolic compartment and induces cell swelling that reduces NFAT5 expression. The addition of ammonium chloride to drinking water of kl/kl mice downregulates the osteo/chondrogenic program elicited by hyperphosphatemia, including Runx2 and TNAP, and reduces vascular calcification (56). Whether a similar strategy might prove useful in the treatment of patients with diabetes, dyslipidemia, or CKD-MBD has yet to be explored. However, these recent data and the impact of matrix stiffness on the osteogenic predilection of vascular progenitors (13) highlight the need to incorporate consideration for the role of mechanobiology in cardiovascular responses to metabolic and nutritional challenges. Emerging bone-vascular interactions are depicted in Figure 1.

FIGURE 1.

Emerging bone-vascular interactions. A bidirectional endocrine relation exists between bone and the vasculature that mutually benefits bone and vascular health. The kidney is an important intermediary in this process via regulation of phosphate excretion (57) and expression of Klotho (58, 59). Importantly, PTH1R signaling maintains bone formation, sustains hematopoietic niche function (60) and endothelial progenitor cell mass (61), promotes intact osteoblast osteopontin 24 and osteocyte FGF23 (58, 62) secretion, supports renal Klotho production (58), and suppresses aortic osteofibrogenic Wnt/β-catenin signaling (40, 48) and vascular calcium accrual (48, 49). PTH1R signaling also reduces aortic (48) and skeletal (63) oxidative stress and maintains the proximity of the microvasculature to the BMU during bone formation (64). Declining renal function and tissue resistance to PTH1R signaling are key features in the perturbation of the bone-vascular axis in the setting of disease. Age-related changes in marrow composition and the vector of bone perfusion may also functionally perturb the bone-vascular axis. In addition, emerging data point to the role of circulating microvesicles—arising from endothelial cells, smooth muscle cells, and formed elements including platelets—in the endocrine regulation of bone-vascular interactions. BMP, bone morphogenic protein; BMU, basic multicellular unit; DMP-1, dentin matrix protein-1; FGF23, fibroblast growth factor 23; PlGF, placental growth factor; PTH, parathyroid hormone; PTHrP, parathyroid hormone related peptide; PTH1R, PTH/PTHrP receptor; RANKL, receptor activator of nuclear factor κB ligand; VEGF, vascular endothelial growth factor; Wnt, wingless-type mouse mammary tumor virus integration site family member. Reproduced from reference 6 with permission.

The Impact of Dietary Calcium and Vitamin D on the Heart and Bone

The vast majority, i.e., >99%, of calcium in the body resides in the skeleton. In the skeleton, increased calcium indicates more bone mineral, because calcium is a constant fraction of hydroxyapatite. Greater bone mass predicts reduced risk of fracture. Thus, stored calcium is a functional reserve. The <1% that is extraskeletal sustains life through a broad array of primary and secondary cell signaling functions and capacity to stabilize proteins. When excess calcium deposition in soft tissue occurs, tissue damage can occur, leading to chronic disease.

Studying calcium metabolism.

Calcium tracers can be used to determine distribution in the body and rates of transfer under different environmental conditions and disease states. Use of calcium tracers in various applications is reviewed by Weaver et al. (65). Calcium has a large number of useful isotopes that provide a wide array of applications. There are 2 radioisotopes, ⁴⁵Ca and ⁴⁷Ca, that are especially useful with animal models, although they both have been used in humans. There are 5 useful stable isotopes: ⁴²Ca, ⁴³Ca, ⁴⁴Ca, ⁴⁶Ca, and ⁴⁸Ca, that can be used safely in any facility and individual. The availability of multiple isotopes allows different tracers to be given orally and intravenously so that complete kinetic analysis is possible. A long-lived isotope, ⁴¹Ca, is also available, and when measured by accelerator mass spectrometry, virtually atom quantities can be measured allowing detection of very early deposition into soft tissues.

Calcium tracers have been particularly useful in elucidating calcium absorption molecular mechanisms (66). Calcium absorption occurs by 2 pathways. Active calcium transport is activated when 1,25-dihydroxyvitamin D elicits transcription of transport proteins including transient receptor potential cation channel subfamily V member 6 (TRPV₆), calbindin D_9k, and PMCA 1b. This pathway is efficient and dominant in conditions of low calcium intake. Calcium absorption during conditions of higher calcium intakes is less dependent on active calcium absorption as the transport proteins become saturated. The non-vitamin D-dependent, passive calcium absorption pathway linearly increases with increased calcium loads over the physiologic range (67).

Factors affecting calcium absorption have also been characterized with calcium tracers. The double calcium isotopic tracer technique, in which one tracer is given orally and another intravenously, is the most sensitive indicator of calcium absorption. Characteristics of an individual such as age and sex steroid hormone deficiency (68) and external factors such as calcium load (69) and food matrix (70) have been identified by using isotopic tracer techniques.

Calcium and vitamin D benefits to bone.

Because calcium is integral to hydroxyapatite and bone continually remodels, a continual supply of calcium in the diet throughout life is essential. Needs are greater during periods of skeletal growth when bone is accruing. Serum calcium is tightly regulated to supply calcium to bone and all tissues. Calcitriol activates calcium absorption, enhances calcium retention, and stimulates osteoclast cells in the bone to release calcium from the bone mineral matrix during periods of low calcium intake, but vitamin D is not required to maintain bone health if calcium intake is sufficient (71). Bone loss by vitamin D receptor-null mice was reversed with dietary calcium and phosphorus (72). This suggests vitamin D is not required for bone health per se but rather is important for maintaining serum calcium. On the other hand, in the NHANES III population-based survey, higher calcium intake was associated (P = 0.005) with higher bone mineral density (BMD), but only in women with lower vitamin D status, i.e., 25-hydroxyvitamin D < 50 nmol/L (73).

The evidence for benefits of calcium and vitamin D supplementation and protection against fracture risk is mixed, likely because of research quality deficiencies such as lack of compliance, studying populations with already Adequate Intake, or use of studies that are too small or too short in duration compared with the development of disease. One meta-analysis reported that calcium and vitamin D supplementation in people aged ≥50 y reduced fracture risk by 12% and in studies with ≥80% compliance, the reduction was 24% (74). In the largest randomized controlled trial, the Women’s Health Initiative, hip BMD was improved with calcium and vitamin D in a subset of the population, but there was no improvement in hip fracture risk (75). However, in women not taking their own supplements who were ≥80% adherent to supplementation for ≥5 y, hip fracture HR was 0.24 (95% CI: 0.07, 0.84) (76). Calcium and vitamin D supplementation is standard therapy for osteoporosis treatment (77) but has not been endorsed by the US Preventative Services Task Force (78). This report has been criticized for not addressing persons with inadequate dietary intake (79). Recommended intakes, bioavailability, and usual intakes of calcium and vitamin D from supplements are addressed elsewhere (80).

The benefit to bone of a wide range of calcium and vitamin D intakes from adulthood through menopause was evaluated in a rat model where diets could be controlled for sufficiently long periods (81). Calcium and vitamin D independently increased tibial and femoral trabecular structures. Vitamin D increased tibial bone width and fracture resistance. Surprisingly few calcium and vitamin D interactions were observed—only for femur length and tibial calcium content. At high calcium and vitamin D intakes, calcium kinetic isotopic tracer analysis showed an increase in the soft tissue compartment. This finding suggests the possibility of soft tissue calcification with prolonged intakes of both calcium and vitamin that exceed recommended levels.

Concern over cardiovascular risks associated with calcium supplementation.

Traditionally, calcium supplementation was associated with only minimal adverse events, i.e., gastrointestinal upset and a small risk of kidney stones (78). Secondary analyses of randomized controlled trials aimed at bone outcome measures showed increased risk of CVD risk factors (82, 83). Fear over risk of heart attacks with supplement use has led to a decline in intake and stimulation of new research to examine the relation of calcium supplementation and CVD risk (84). The controversy over whether to recommend supplementation to reduce risk of osteoporosis or to not recommend it for fear of cardiovascular risk or lack of benefit to bone is not entirely resolved. No professional society or policy body has yet taken the stance that there is sufficient evidence to warn against supplementation to lower cardiovascular risk.

Epidemiologic studies in this area have been mixed. Michaëlsson et al. (85) found a U-shaped relation between calcium intake and cardiovascular mortality in a large Swedish cohort of women, particularly indicating increased risk in women who used calcium supplements on top of already high calcium intake (∼1400 mg/d). On the other hand, Van Hemelrijck et al. (86) found no association between dietary or supplemental calcium intake and cardiovascular mortality in an analysis of NHANES III. The evidence base suggesting a risk of CVD with calcium supplementation has suffered from lack of a clear underlying mechanism for calcium intake, no consistent dose-response relation, and insensitive methods to assess soft tissue calcification or accuracy and consistency of cardiovascular events. To help address these knowledge gaps, a study was conducted to specifically evaluate the impact of chronic high calcium intake on soft tissue calcification in an animal model relevant to humans by using a novel approach to determine early soft tissue calcification. The Ossabaw miniature swine fed an atherogenic diet developed metabolic syndrome with associated soft tissue calcification and CVD. A sensitive approach to measuring early soft tissue calcification was developed by using ⁴¹Ca administration and subsequently analyzing ⁴¹Ca:Ca ratios in coronary arteries. Coronary artery accumulation of ⁴¹Ca:Ca in pigs with metabolic syndrome was much higher than in healthy pigs. This approach was validated against percentage of wall coverage by ultrasound (Figure 2) and is capable of detecting earlier shifts in calcification (87).

FIGURE 2.

Coronary artery calcium uptake and level of atherosclerosis in Ossabaw miniature swine. (A) ⁴¹Ca:Ca ratio of right coronary artery samples of lean (n = 7) and MetS (n = 7) pigs (means ± SDs, P < 0.05). Values for 6 lean pigs and 2 MetS pigs were below the detection limit; the upper limit of detection was used for the pigs for calculating the means. (B) Coronary atheroma assessed by intravascular ultrasound. Percent wall coverage was obtained by pullback of the ultrasound catheter along the length of the coronary artery. Ca, calcium; MetS, metabolic syndrome. Adapted from reference 79 with permission.

To test the effect of high compared with recommended calcium intake, adult Ossabaw miniature swine were randomly assigned to high-calcium diets (2%, equivalent to the upper level in humans) from dairy or calcium carbonate and control diets (0.5%, recommended requirement for pigs) and fed for 6 mo until soft tissue calcium deposition had begun. High calcium intakes from neither source resulted in significant differences from control pigs in coronary artery ⁴¹Ca deposition (P > 0.05) or any traditional measure of cardiovascular function, plaque wall coverage, stroke volume and ejection fraction, or histologic assessment of calcification (88). Thus, this study suggests that calcium from diet or supplements up to the upper level does not appear to increase risk of CVD.

Calcium and vitamin D are essential nutrients. Current evidence supports that it is prudent to follow the Dietary Guidelines for Americans (89) and calcium and vitamin D recommendations from the Institute of Medicine (90) and to aim for intakes around the RDA but below the tolerable upper level. Whether or not calcium and vitamin D supplementation is beneficial to bones depends on whether an individual is deficient and other factors. Based on the study in pigs, calcium intakes up to the tolerable upper level are unlikely to increase risk of CVD. One group that remains vulnerable to bone and cardiovascular disorders is those with CKD, and this population requires further study. The impact of diet is not completely understood, but a study of calcium carbonate supplementation to control phosphorus with the use of calcium tracer kinetics suggested soft-tissue deposition may be occurring in patients with CKD (91).

The Role of Dietary Phosphorus in Cardiovascular and Bone Health

Phosphorus is an essential nutrient with a variety of functions in the body, including structure as a component of bone mineral, cell membranes, and nucleic acids; acid-base balance as an intracellular buffer and titratable acid buffer; and in energy metabolism as part of the compounds adenosine triphosphate, guanosine triphosphate, etc. The majority (∼85%) of the body’s phosphorus is found in bone mineral with calcium to form hydroxyapatite [Ca₁₀(PO₄)₆(OH)₂] (92). Thus, phosphorus deficiency manifests in osteomalacia or rickets. However, dietary phosphorus deficiency is rare because of its widespread presence in the food supply; it is found naturally in protein foods like meat, dairy, nuts, and seeds and also in grains and other plant foods. In addition, inorganic phosphates, which are highly absorbed, are added to many foods for a variety of purposes including emulsification, leavening, pH control, thickening, and increasing the shelf-life of products (93). Subsequently, concern has arisen over potential consequences of excessive intake of phosphorus on bone and cardiovascular health, particularly for those with CKD. In CKD, abnormal phosphate metabolism is central to the development of CKD-MBD, which is characterized by interrelated laboratory abnormalities in mineral metabolism, bone disease, and vascular calcification (44, 94). As a consequence of CKD-MBD, patients with end-stage renal disease have increased risk of cardiovascular mortality compared with the general population (95) and more than a 4-fold risk of fracture (96).

The ratio of calcium to phosphorus in the diet has been investigated related to bone health, with the hypothesis that the relative intakes of these nutrients determine their adequacy or excess. Optimal ratios have been estimated as 2:1 during infancy and 1.3–1.5:1 beyond. However, calcium balance studies in adults studied under a wide range of calcium:phosphorus conditions from 0.08:1 (>12× phosphorus compared with calcium) to 2.4:1 show no effect of ratio on calcium balance (97–99). High phosphorus stimulates PTH, and restriction of dietary phosphorus can lessen or correct secondary hyperparathyroidism. Thus, high dietary and serum phosphorus can influence bone disease, particularly in patients with CKD and end-stage renal disease through detrimental effects of high PTH on bone (100).

Relation between phosphorus and cardiovascular disease and mortality.

Serum phosphate has been associated with CVD and mortality consistently across numerous studies that included healthy people (101, 102), people with pre-existing heart disease (103), and patients with CKD (104, 105). It is notable that risk for CVD has been observed to increase even over a range of normal serum phosphate (102). Few studies, to our knowledge, have investigated the relation between dietary phosphorus intake and CVD and mortality, partially because of the challenges of assessing dietary intake (106). Chang et al. (107) found increased risk of all-cause mortality with high dietary phosphorus intakes in an analysis of NHANES III. Increased risk was reported to start at intake levels of ∼1400 mg/d; however, confidence intervals suggest that risk may not increase until high intakes are reached. Interestingly, another study of people with reduced renal function (estimated glomerular filtration rate) in NHANES III showed no relation between dietary phosphorus intake level and mortality (108).

One mechanism by which phosphorus may influence CVD and mortality is through the actions of FGF23 on left ventricular mass (43), as described above. Thus, it is plausible that high dietary phosphorus may affect cardiovascular risk indirectly through FGF23 effects. Supportive of this concept, Yamamoto et al. (109) have shown an association between higher dietary phosphorus intake and greater left ventricular mass in women, but not men. In addition, the direct effect of phosphate on VSMC transdifferentiation to osteoblastic-like cells capable of laying down collagen matrix and mineral is another plausible mechanism by which dietary phosphorus may affect CVD (110). Supportive of this, a study in CKD rats showed that restricting dietary phosphorus prevents the vascular calcification that is otherwise observed in rats fed a normal phosphorus diet (111). These associational and mechanistic studies indicate there is good reason for caution regarding excessive dietary phosphorus intake in the general population and particularly in special groups like people with CKD. However, further research is needed for more definitive recommendations regarding the potential dangers of phosphorus excess.

Racial Differences and the Impact of Nutrition in Osteoporosis and Cardiovascular Disease

Race and ethnicity.

Race and ethnicity are 2 different but related constructs that have a tremendous influence on the incidence of chronic diseases (112). Race encompasses biological/genetic characteristics, whereas ethnicity describes identification with cultural practices that include but are not limited to geographic region, including language, heritage, religion, tradition, and customs. Health disparities research investigates differences in health across disadvantaged populations including factors of race and ethnicity and determines whether they are attributable to decreased access to necessary health care services or caused by other factors such as greater occurrence of behavioral risk factors for a disease or condition (113). Differences in screening, treatment, adherence, and management of chronic diseases play an important role on health disparity between races. A complex framework exists for understanding the multitude of factors that exist and interact to cause health disparities (112). This section summarizes the role of nutrition in racial disparities with regard to osteoporosis and CVD.

Race disparity and bone health.

Low bone mass and osteoporosis greatly increase the susceptibility to bone fracture. The risk of osteoporosis and bone fracture increases linearly with age. However, several other factors are also related to bone disease including sex, race/ethnicity, and nutrition. Women carry a disproportionate burden of bone disorders related to aging; in fact, 1 in 3 women over the age of 50 y will experience a bone fracture in her lifetime (114).

According to national estimates from adults over the age of 50 y from NHANES 2005–2010, standardized to the population by using census data, ∼10% of the US population (10.2 million) and ∼44% (43.4 million) older adults have osteoporosis and low bone mass, respectively. However, differences exist in the prevalence of bone disorders among women by race (115). Approximately 7.7 million non-Hispanic white adults have osteoporosis compared with 0.5 million and 0.6 million non-Hispanic black and Mexican American adults, respectively (115). More importantly than prevalence estimates, non-Hispanic white women tend to have more hip fractures than women of other races/ethnicities (115). Another NHANES 2007–2008 analysis reveals that despite lower levels of physical activity, the prevalence of osteoporosis remains much lower in non-Hispanic black and Mexican Americans (n = 2819, aged 40–80 y than non-Hispanic white adults (116).

Calcium and vitamin D have a well-established role with regard to bone health; however, the exact mechanisms by which these nutrients interact with race are largely unknown. Across most age and race/ethnic groups, intakes of calcium and vitamin D are very low and well below recommendations (117, 118). The proportion of non-Hispanic black Americans with calcium and vitamin D intakes less than the estimated requirement was higher than all other race/ethnic groups in the United States (119). Differential intakes of milk and dairy products exist between race/ethnic groups and may be related to lactose tolerance; however, this relation is not well characterized, and a standard diagnostic criteria for lactose tolerance does not exist (120). Interestingly, despite lower intakes of bone relevant micronutrients and dairy in the diet, non-Hispanic black women have the lowest incidence of bone fracture of all the observed race/ethnic groups and higher BMD than non-Hispanic white females. Studies have shown that non-Hispanic blacks have lower calcium excretion and higher calcium retention (121, 122) and achieve 5–15% higher peak bone mass than non-Hispanic whites. Serum 25-hydroxyvitamin D concentrations are also lower among non-Hispanic black Americans than in other racial groups (123). The levels of vitamin D-binding proteins are stable (124–126) and do not change with the change in the levels of 25-hydroxyvitamin D, but the levels of vitamin D-binding proteins are relatively lower in non-Hispanic blacks than non-Hispanic whites (123, 127). PTH is one of the major hormones that influences bone resorption and calcium absorption in the intestine and impacts BMD. Studies have shown that the levels of PTH differ between race/ethnic groups and may be one of the possible mechanisms for the differences in bone health between race/ethnic groups (128). Asian-American females have similar BMD to non-Hispanic white women; however, they experience fewer falls and fractures (129). Thus, BMD alone may not be the most important component in understanding how race influences bone health and fracture risk. To our knowledge, very little is known about the bone health of Native Americans and Hispanics aside from Mexican Americans in the United States. In summary, racial differences in BMD and fracture risk in non-Hispanic black women appear to be independent of physical activity, PTH, and dietary intakes. Given that more women will experience a bone fracture than breast cancer, myocardial infarction, and coronary death in a given year, characterizing how race and ethnicity impact bone health and disease risk is critical.

Race disparity and CVD.

Obesity, being overweight, and diabetes mellitus are other relevant major health conditions in which race disparity is very profound. Analyses of data from NHANES 2009–2010 showed that non-Hispanic whites have the lowest rates of being overweight, while non-Hispanic blacks have the highest rates of obesity, with >13% having a BMI (in kg/m²) of ≥40 as compared with 6% among non-Hispanic whites (130). Weight status and diabetes are strong, independent risk factors for CVD. CVD is the leading cause of mortality among adults 65 y and older (131). Among types of CVD, coronary heart disease accounts for ∼48% of the deaths followed by stroke (16%) (131). Non-Hispanic black men and women have the highest prevalence of high blood pressure, myocardial infarction, stroke/transient ischemic attack, as well as death from these diseases when compared with other race/ethnic groups in the United States across multiple years of NHANES, whereas Mexican American men and women have the highest prevalence of elevated cholesterol (≥200 mg/dL) (132).

Diet and race interactions.

Diet and dietary patterns play an important role in the health of an individual and can be modifiable causes that contribute to race/ethnic disparities. Understanding how dietary differences interact with race is critical (133). Data from a multiethnic cohort (cohort includes >215,000 participants, including African American, Native Hawaiian, Japanese America, Latino, and Caucasian men and women, living in Hawaii and Los Angeles in 1993–1996) have demonstrated a complex interaction by race and sex for diet and health outcomes (134). For example, higher intakes of vegetables substantially reduced the risk of fatal stroke in African American women, whereas among Japanese women higher fruit intake and lower meat intake was associated with reduced risk of fatal stroke (134). It is interesting to note in this cohort that no significant associations by race were observed for males (P > 0.05).

Understanding the relation between diet and nutrition status, race, and chronic disease will provide insights to reduce health disparities. Health professionals need to understand that race and ethnicity are factors that should be considered in providing dietary advice to prevent, manage, or treat diseases. In public health, “cultural competency” refers to the attitudes, behaviors, and policies that guide health care systems, agencies, and personnel to work effectively in cross-cultural situations. Furthermore, cultural competency extends to include the ability of health care providers and systems to care for diverse patients in a manner that meets their social, cultural, and linguistic needs. Many more multidimensional studies are needed to identify appropriate interventions to reduce race disparity in health.

Conclusions

Cardiovascular and skeletal health are connected through interacting hormonal and cellular processes, and CKD presents conditions in which these relations are amplified and the risk of CVD and bone disease is greatly increased. Current evidence for the role of calcium, vitamin D, and phosphorus in the development of CVD and osteoporosis points toward a conservative approach of aiming for intakes around current recommended levels but cautions against excessive intakes, particularly in those with compromised renal function. It seems prudent to avoid excessive phosphorus intake that is facilitated through widespread natural sources and food additive use, as well as to avoid excessive calcium intake through high-dose supplements. Racial and ethnic disparities in the prevalence of CVD and osteoporosis exist, and there is evidence of race/ethnicity-nutrition interactions in disease susceptibility. Further investigation is needed to understand the role of nutrition in reducing these disparities.