Vitamin D and Atherosclerotic Cardiovascular Disease

Abstract

Context:

A large body of experimental and observational data has implicated vitamin D deficiency in the development of cardiovascular disease. However, evidence to support routine vitamin D supplementation to prevent or treat cardiovascular disease is lacking.

Evidence Acquisition:

A comprehensive literature review was performed using Pubmed and other literature search engines.

Evidence Synthesis:

Mounting epidemiological evidence and data from Mendelian randomization studies support a link between vitamin D deficiency and adverse cardiovascular health outcomes, but randomized trial evidence to support vitamin D supplementation is sparse. Current public health guidelines restrict vitamin D intake recommendations to the maintenance of bone health and prevention of fractures. Two recently published large trials (VITAL and ViDA) that assessed the role of moderate-to-high dose vitamin D supplementation as primary prevention for cardiovascular outcomes in the general population had null results, and previous randomized trials have also been generally negative. These findings from general population cohorts that are largely replete in vitamin D may not be applicable to chronic kidney disease (CKD) populations, in which the use of active (1α-hydroxylated) vitamin D compounds is prevalent, or to other high-risk populations. Additionally, recent trials in the CKD population, and trials using vitamin D analogues have been limited.

Conclusions:

Current randomized trials of vitamin D supplementation do not support benefits for cardiovascular health, but the evidence remains inconclusive. Additional randomized trials assessing larger numbers of participants with low baseline vitamin D levels, having longer follow-up periods, and testing higher vitamin D dosages, are needed to guide clinical practice.

INTRODUCTION

Cardiovascular disease (CVD) is one of the leading non-communicable causes of mortality(1), causing one third of deaths worldwide. Global health studies have revealed an epidemiologic transition from infectious to non-communicable diseases such as CVD as the predominant cause of death in industrialized countries.(2) This trend is also becoming apparent in low and middle-income countries, commonly affecting individuals at premature ages or from underserved ethnic groups or regions.(3,4) Atherosclerotic CVD (ASCVD), including coronary heart disease and stroke, tends to be a chronic inflammatory and thrombo-occlusive condition affecting medium-sized and large arteries. (5) ASCVD in regions such as Sub-Saharan Africa is now an established major cause of death, albeit largely secondary to non-atherosclerotic conditions such as rheumatic heart disease, HIV-related disorders and hypertensive heart disease.(6) In industrialized countries and regions where population aging has evolved such as in South America, unhealthy lifestyles have become increasingly prevalent accounting for higher rates of ASCVD.(4,7) It is therefore not surprising that CVD represents a significant healthcare economic burden worldwide.

CVD burden has been attributed to certain behaviors such sedentary lifestyle, unhealthy dietary patterns, smoking and excessive alcohol consumption.(3) Growing evidence supports significant cause-effect relationships of most of these factors with CVD.(8,9) Other environmental factors such as air pollution and conditions such as inflammatory arthropathies, HIV, diabetes and chronic kidney disease (CKD) are now known to be important risk factors for CVD. A considerable body of evidence has also implicated hypovitaminosis D as an important risk for CVD.

Low blood levels of vitamin D have been linked to a number of adverse health outcomes including all-cause mortality, cardiovascular disease as well as reduced bone density and fracture risk, metabolic syndrome, malignancy, autoimmune conditions and infection.(10) Overt vitamin D deficiency results in the striking phenotype of rickets in children, and osteomalacia in adults and children. It has therefore been intensively studied for almost a century. Animal models have shown that vitamin D receptor (VDR) knockout mice develop overt CVD, including upregulation of the renin-angiotensin-aldosterone (RAAS) system, hypertension, left ventricular hypertrophy and heart failure.(11,12) In humans, a large meta-analysis of observational studies (n=849 412) found that low concentrations of 25-hydroxyvitamin D (25(OH)D) were associated with an increased risk of cardiovascular as well as cancer, and non-cardiovascular non-cancer deaths.(13) Each 10 nmol/L reduction in 25(OH)D was associated with a 16% increase in all-cause mortality; the pooled relative risk (RR) was 1.35 (95% CI 1.13 to 1.61) for death from cardiovascular disease, 1.14 (1.01 to 1.29) for death from cancer, 1.30 (1.07 to 1.59) for non-vascular, non-cancer death, and 1.35 (1.22 to 1.49) for all-cause mortality.(13) These data were supported by findings from a Scandinavian Mendelian randomization study (n=95,766) where a genetically determined difference in circulating 25-hydroxyvitamin D (25(OH)D) concentrations of 20 nmol/L was associated with a significantly increased risk of all-cause, cardiovascular and cancer mortality (HR 1.19, 1.18 and 1.12 respectively),(14) although other genomic studies have been inconsistent. Given the considerable proportion of the general population at risk for vitamin D inadequacy, establishing the implications for cardiovascular health is of high importance: the National Health and Nutrition Examination Survey (NHANES) for 2007 through 2010 revealed that 13% of Americans aged 1 to 70 years of age are at risk for vitamin D inadequacy.(15) Additionally, NHANES identified a marked decrease from 45% (43–47%) to 23% (20–26%) in serum 25(OH)D levels of 30 ng/ml or more between NHANES III (1988–1994) and NHANES IV (2001–2004).(16) However, some of these changes may be attributed to fluctuations in the DiaSorin RIA assay kit used.(17)

Serum 25(OH)D is commonly measured, and low levels are often treated in clinical practice with over-the-counter supplements.(10) National and international guidelines set thresholds for deficiency, insufficiency and repletion (based mostly on skeletal outcomes), and recommend daily intakes for the general population.(18) In kidney disease, vitamin D has been used for more than 6 decades to treat renal bone disease,(19) though mostly by using synthetic vitamin D compounds rather than naturally occurring ones. Despite this, there is a lack of data on the effect of such practices on meaningful clinical outcomes over and above biochemical parameters such as parathyroid hormone (PTH) and calcium.(20) The 2011 dietary reference (DRI) intake report by the Institute of Medicine (IOM) concluded that bone health was the only outcome whereby causality was established by available evidence.(21) Using a risk assessment framework, the IOM committee determined that for CVD, cancer, diabetes, falls, physical performance, autoimmune disorders and other extraskeletal chronic disease outcomes, the evidence was deemed to be inconsistent, inconclusive as to causality and insufficient to serve as a basis for DRI development.

This review will consider contemporary data on vitamin D and their application to clinical practice for the management of ASCVD. We conducted a comprehensive literature review considering the biological mechanisms and current experimental, clinical and randomized trial data available to guide vitamin D therapy in cardiovascular disease prevention and/or treatment. PubMed and other literature search engines were utilized.

Vitamin D nomenclature and physiology

Vitamin D is a generic term that refers to a group of compounds derived from 7-dehydrocholesterol (7DHC) (Figure 1). In this review however, this term is exclusively applied to cholecalciferol (vitamin D₃), formed through dermal photoconversion of 7DHC. Cholecalciferol is biologically inert but is converted to compounds that bind to the nuclear vitamin D receptor. From skin, cholecalciferol enters the circulation, where it has a half-life of ~24h due to substrate-dependent hepatic conversion to 25-hydroxyvitamin D (25(OH)D). 25(OH)D is the dominant circulating vitamin D metabolite, has a long half-life of ~3 weeks, and is used to determine vitamin D status given that its concentrations reflect supply, metabolism and catabolism over time and low concentrations are ubiquitous in vitamin D deficiency rickets or osteomalacia.(22) 25(OH)D is converted to 1,25-dihydroxyvitamin D (1,25(OH)₂D, calcitriol) by the 1α-hydroxylase (CYP27B1). 1,25(OH)₂D is a calcitropic steroid hormone with a short half-life of only 4–6 hours and a high affinity for the nuclear vitamin D receptor (VDR).

Figure 1: Vitamin D compounds and synthesis:

Choleclaciferol is formed through dermal photoconversion of 7-dehydrocholesterol (7DHC). Cholecalciferol is biologically inert, but is converted by various enzymes to compounds that bind the nuclear vitamin D receptor (VDR). 1,25-dihydroxyvitamin D (1,25(OH)₂D, calcitriol) is the active form of vitamin D that is synthesized from 25(OH) ₂D by 1α-hydroxylase (CYP27B1), now known to be expressed in almost all human tissues. 1,25(OH) ₂D has a high affinity for the VDR.

1α-hydroxylase was initially identified in proximal tubular cells of the kidney, but is now known to be expressed throughout the nephron including the distal tube and collecting duct.(23) Renal 1α-hydroxylase appears to account for the majority of circulating 1,25(OH)₂D that mediates endocrine (or classical) effects. Outside of the kidney, 1α-hydroxylase has been found to be expressed in almost all human tissue types, including the skin (basal keratinocytes, hair follicles), lymph nodes (granulomata), colon (epithelial cells and parasympathetic ganglia), brain (cerebellum and cerebral cortex), prostate, breast, testes, and placenta (decidual and trophoblastic cells), pancreas (islets), adrenal medulla, and notably in myocardium and vascular cells (endothelial and vascular smooth muscle cells).(24) Extra-renal synthesis of vitamin D mediates autocrine/paracrine or “non-classical” functions of vitamin D. It is unclear how much extra-renal 1α-hydroxylase activity contributes to circulating vitamin D levels. However, anephric individuals who were administered inactive vitamin D₃ have been shown to have detectable serum 1,25(OH)₂D levels suggesting significant extrarenal 1α-hydroxylase activity.(25)

Similar to 1α-hydroxylase, the VDR is ubiquitously expressed and this explains the multi-system effects observed during hypovitaminosis D.(26) Vitamin D deficiency results in notable mineral-bone abnormalities, including reduced intestinal calcium and phosphate absorption, increased parathyroid hormone (PTH) concentrations and reduced bone mineral density,(10,26) features recapitulated in VDR^−/− mice.(27) This is not surprising given that 1,25(OH)₂D levels are tightly regulated by PTH, serum calcium, phosphorus and the phosphatonin, fibroblast growth factor (FGF)-23. Dietary calcium can exert direct regulatory effects on the 1α-hydroxylase enzyme through modulations in serum calcium levels. Both dietary calcium and phosphate indirectly regulate the 1α-hydroxylase enzyme by modulating PTH production at the parathyroid gland. Low calcium and phosphate concentrations lead to the stimulation of 1α-hydroxylase to produce active 1,25(OH)₂D.

FGF-23 is a bone-derived hormone that directly suppresses 1α-hydroxylase activity at the kidney and its function is dependent upon expression of a specialized functional receptor at the kidney, composed of fibroblast growth factor receptor (FGFR)-1 and the transmembrane α-Klotho protein.(28,29) High FGF-23 levels lead to suppressed of 1α-hydroxylase activity and consequently, reduced 1,25(OH)₂D production. Given that α-Klotho is widely expressed in extrarenal tissues in humans including arterial, epithelial, endocrine, reproductive and neuronal tissues,(30) it is possible that FGF-23 can regulate extrarenal 1,25(OH)₂D synthesis in α-Klotho producing tissues. High 1,25(OH)₂D concentrations reduces 1α-hydroxylase enzyme activity via inhibition of PTH production, forming a negative feedback loop to prevent the development of hypervitaminosis D. Additionally, upregulation of the 24-hydroxylase enzyme by high 1,25(OH)₂D stimulates the catabolic pathway resulting in reduced 1,25(OH)₂D concentrations.

MOLECULAR EFFECTS ON THE ATHEROGENIC PROCESS

Robust experimental studies have shown strong links between vitamin D deficiency and pathogenic processes underlying the development of cardiac failure, such as endothelial dysfunction, reduced coronary flow, and subclinical atherosclerosis in patients with normal, or near-normal coronary arteries.(11,31) Signalling components of the vitamin D hormonal system are widely expressed across various cell types of the cardiovascular system, including myocardial cells, endothelial cells, vascular smooth muscle cells (VSMCs), fibroblasts and pericytes. This reflects the critical role of vitamin D in the regulation of cardiovascular health, particularly considering the role that these cell types, in addition to immune cells, play in the development of atherosclerosis. The presence of vitamin D signalling components and a functional 1α-hydroxylase enzyme in multiple cardiovascular cell types reflect an interplay between local production to exert autocrine/paracrine effects and response to endocrine effects in the regulation of normal cardiovascular physiology.(32)

VDR knockout mice exhibit prominent cardiovascular dysfunction, including upregulation of the renin-angiotensin-aldosterone (RAAS) system, hypertension, left ventricular hypertrophy and heart failure in addition to changes resembling premature aging such as alopecia, abnormal blood mineral levels, hyperparathyroidism, and defective T cell and macrophage function.(33,34) Conversely, repletion of vitamin D has been shown to reverse these changes in mice.(35) Significantly, VDR knockout mice develop multi-organ thrombus formation after administration of lipopolysaccharide.(36) These mice were found to have enhanced platelet aggregation, reduced nitric oxide (NO) synthase expression, and reduced expression of anti-thrombin in the liver, and thrombomodulin in the aorta, liver and kidney, thereby promoting the atherogenic process. Vitamin D exerts pleiotropic anti-atherogenic effects, such as regulation of vascular cell growth, migration and differentiation, immune response modulation, and regulation of cytokine production, inflammation and fibrotic pathways all of which play a critical role in atherosclerotic plaque formation and rupture.(37) Potential mechanisms by which VDR activation can modulate the development of cardiovascular disease is summarized in figure 2.

Figure 2: Vitamin D receptor (VDR) activation and the atherogenic process:

VDR activation can exert pleiotropic cardiovasculo-protective effects through multiple mechanisms. VSMCs (Vascular smooth muscle cells); VEGF (vascular endothelial growth factor).

Effects on vascular cells

VDR activation in endothelial cells (ECs) can modulate vascular tone through various mechanisms: Vitamin D stimulates nitric oxide (NO) production by activating NO synthase,(38,39) regulates the release of vasoconstrictor metabolites of arachidonic acid(38,40) and downregulates expression of cyclooxygenase-1, the major source of endothelium-derived contracting factors(41) By reducing calcium influx into ECs, VDR activation can acutely modulate vascular tone.(42) Vitamin D exerts anti-oxidative effects in ECs, in vitro by suppressing NADPH oxidase, leading to reduced superoxide synthesis(43) and blocking the extrinsic caspase cascade.(44) Importantly, vitamin D can inhibit intercellular adhesion molecule-1 and vascular cell adhesion molecule-1 in ECs in vitro, factors that are involved in subendothelial migration of monocytes and subsequent transformation into foam cells.(45,46)

Vascular smooth muscle cells (VSMCs) form the medial layer of the arterial wall and play a pivotal role in pathogenesis of atherosclerosis through proliferation and migration from the media to the intima, morphological changes and secretion of inflammatory factors.(47) Vitamin D has been shown to stimulate antiproliferative effects of VSMCs via inhibition of epidermal growth factor and endothelin(48,49), and regulate acute influx of Ca²⁺ into the cell, in vitro.(50) High doses of 1,25(OH)₂D have been shown to induce VSMC migration while physiological doses of 25(OH)D and 1,25(OH)₂D inhibits migration and proliferation.(51–53) Activation of VDRs increases elastogenesis and stabilization of the VSMC layer by stimulating production of myosin, collagen type 1, matrix metalloproteinase-9 and elastin.(54) By stimulating prostacyclin production via the cyclooxygenase pathway in VSMCs in vitro, 1,25(OH)₂D may inhibit aggravation of atherosclerosis.(55) Activated vitamin D analogs have been shown to suppress plasminogen activator inhibitor-1, a pro-thrombotic and pro-atherogenic factor in coronary artery smooth muscle cells, in vitro.(56)

Vascular calcification occurs in healthy aging, in conjunction with atherosclerosis (intimal calcification) and at an accelerated rate in CKD (intimal and medial calcification), where it is strongly associated with cardiovascular events and mortality. Vitamin D deficiency has been associated with increased arterial calcification across a number of studies,(57–60) and with the development of coronary artery calcification.(61) Vitamin D can promote the production of vitamin K-dependent proteins and exert synergistic effects. This interaction appears critical given that vitamin K deficiency is associated with vascular calcification.(62) Experimental studies suggest that the role of VDRAs in the regulation of vascular calcification is dependent on the specific analogue. In animal models, administration of pharmacological doses of 1,25(OH)₂D resulted in increased aortic calcification, however this was not seen in animals treated with the synthetic 1,25(OH)₂D analogue 19-nor-1,25-dihydroxyvitamin D2 (paricalcitol).(63) In one model, both 1,25(OH)₂D and paricalcitol were protective against vascular calcification at dosages sufficient to correct secondary hyperparathyroidism.(64)

Administration of active vitamin D sterols result in a reduction of plasma vitamin D concentrations, due to potent induction of the vitamin D catabolic pathway, in vivo.(58) These findings are consistent with evidence that polymorphisms in CYP24A1 that result in lower 25(OH)D concentrations are associated with increased coronary artery calcification.(65) Ensuring vitamin D repletion may therefore allow the autocrine and paracrine synthesis of 1,25(OH)₂D in cardiovascular tissues, activating VDR-dependent calcification inhibitory pathways. 1,25(OH)₂D and its analogues, administered systemically may overwhelm the autoregulatory system by potently inducing 24-hydroxylase expression and potentially result in local vitamin D deficiency at the tissue level.

Regulation of immune cells

The atherogenic process involves both the innate and adaptive immune system. Vitamin D is carried by low-density lipoprotein particles and is internalized by various cells, such as monocytes that contain a functional 1α-hydroxylase enzyme. Monocytes migrate trans-endothelially and are transformed into foam cells. Local synthesis of 1,25(OH)₂D can decrease expression of monocyte adhesion expression and alter macrophaqe gene expression, facilitate cholesterol egress, lower acetylated and oxidized LDL uptake, and reduce foam cell production(66,67) and influence the atherosclerotic process by regulating endothelial cells and VSMCs (e.g. type I collagen, VEGF production, matrix metalloproteinases). VDR activation results in inhibition of interferon-ϒ and upregulation of IL-10, vitro.(68,69) Additionally, VDRs inhibit IL-1β and IL-6 production preventing macrophage activation and plaque instability.(70) Stimulation of IL-4 synthesis by VDR activation promote anti-atherogenic properties of Th2 cells in vitro, which secrete anti-atherogenic cytokines such as IL-5, IL-10 and IL-13.(71) These changes, seem likely to underlie some of the vasculo-protective properties of VDR activators.

Indirect effects of vitamin D on atheroma formation

VDR activators can protect against atherogenic formation indirectly by regulating other systemic processes. Vitamin D has been shown to regulate the renin angiotensin aldosterone (RAAS), a key regulator of blood pressure. Both angiotensin II and aldosterone are also involved in the atherogenic process(72) and this is promoted by vitamin D deficiency leading to activation of the RAAS system.(26) Conversely, VDR activators can suppress renin and angiotensinogen gene expression by blocking the NF-kB pathway.(11,73) There is also growing evidence that vitamin D can regulate β-cell function and insulin sensitivity(74) and that deficiency is associated with dyslipidemia,(75) both of which can promote atheroma formation. However, a detailed discussion of this is beyond the scope of this article.

EPIDEMIOLOGICAL AND INTERVENTIONAL DATA IN THE GENERAL POPULATION

Many studies have demonstrated efficacy of cholecalciferol, alone or with calcium supplements, in reducing osteoporotic fracture risk, and calcium and vitamin D have been integral to the treatment of osteoporosis for several decades, as remains the case today.(76) A considerable body of clinical and epidemiologic data suggests a role for vitamin D in reducing cardiovascular disease and mortality. Observational studies have shown that vitamin D deficiency is associated with hypertension, LVH, overt coronary heart disease and myocardial infarction, and is prognostic for major post-infarction adverse events including, heart failure hospitalizations, recurrent acute myocardial infarction and death(77–80). In a meta-analysis of observational studies including 65 994 participants, Wang et al showed that vitamin D deficiency was associated with an increased risk of cardiovascular disease, cardiovascular mortality, myocardial infarction and stroke.(12) Similarly, Chowdhury et al demonstrated a significant association of low vitamin D concentrations with cardiovascular death in a meta-analysis of observational studies including 849 412 participants comparing the lower to the upper vitamin D tertile (pooled relative risk 1.35, 95%CI 1.13 to 1.61).(13) In another meta-analysis that investigated the relationship between vitamin D and carotid atherosclerosis, vitamin D concentrations were inversely associated with carotid atherosclerosis.(81)

Findings from Mendelian Randomization (MR) studies have been inconsistent. Aspelund et al reported a HR of of 1.20 (95% CI 1.15–1.25) for all-cause mortality of a 20nmol/L difference in 25(OH)D in 10,501 patients.(82) In a large study of 95,766 Danish patients, the same reduction in 25(OH)D was associated with increased all-cause and cancer mortality, but not with increased cardiovascular mortality.(14) This lack of association with cardiovascular mortality is consistent with a recent report by Brondum-Jacobsen et al, where genetic determinants of 25(OH)D were not significantly associated with atherosclerotic ischemic heart disease (HR 0.98, 95% CI 0.76–1.26) or myocardial infarction (HR 1.15 (95% CI 0.83–1.59) in 92,416 Danish participants.(83) Other smaller MR studies have failed to show an association of genetic determinants of circulating 25(OH)D concentrations with all-cause, cardiovascular or noncardiovascular deaths(84) or diabetic microvascular complications (DMI).(85) A small but statistically significant reduction in systolic BP of 0.37mmHg was found to be associated with a genetically determined increment of 10 nmol/l in circulating 1,25(OH)₂D in a large mendelian randomisation study (n=49,363).(86) However, the Women’s Genome Health Study (WGHS) that included 23,294 women of European ancestry did not find any association between 5 selected vitamin D-related single nucleotide polymorphisms (SNPs) with systolic or diastolic blood pressure.(87) It is noteworthy that there is some evidence that the association of vitamin D deficiency with health outcomes may in turn vary with common vitamin D receptor polymorphisms, and this may complicate interpretation of MR studies.(88) Further, polymorphisms investigated in MR studies vary, and may not include all genetic variants driving circulating 25(OH)D concentration.(89)

Risk factors for ASCVD such as dyslipidemia, hypertension and obesity, features that capitulate the metabolic syndrome significantly increase the risk of ASCVD. Several cross-sectional studies have shown that high serum 25(OH)D levels are associated with a favourable serum lipid profile.(90–92) However, clinical trials have showed conflicting results as discussed in further detail below.

Treatment with vitamin D in the general population

A number of meta-analyses have examined vitamin D replacement therapy with all-cause mortality and cardiovascular outcomes. A meta-analysis of randomized trials including 13,637 participants randomized to cholecalciferol or placebo reported that supplementation with cholecalciferol in subjects without kidney failure reduced all-cause mortality (HR 0.89, 95% CI 0.80 – 0.99).(13) Interestingly, there was no benefit with ergocalciferol. This is consistent with reports indicating that greater increments in total and free 25(OH)D are achieved with equivalent doses of cholecalciferol compared to ergocalciferol.(93,94) The benefit associated with cholecalciferol may be mediated by its effects on a variety of established cardiovascular risk factors. Mirhosseini et al reported a meta-analysis including data from 81 randomised trials, demonstrating that randomization to vitamin D resulted in significant reductions in systolic blood pressure, diastolic blood pressure, serum PTH, hs-CRP, total cholesterol, LDL, triglycerides and a significant increase in HDL.(95) In contrast to the reported effect on blood pressure, the Women’s Health Initiative trial failed to identify any reduction in SBP (0.22 mmHg, 95% CI −0.05 to 0.49) or DBP (0.11 mmHg, 95% CI −0.04 to 0.27) in 36,282 post-menopausal women randomised to elemental calcium and cholecalciferol or placebo with a median follow-up of 7 years.(96) The investigators also found no reduction in coronary or cerebrovascular risk in the treatment arm.(97) However, this trial used a very low daily dose of cholecalciferol (400IU) in combination with 1000 mg of elemental calcium. Additionally, in a meta-analysis by Elamin, et al that examined 51 trials, vitamin D supplementation was not associated with changes in diastolic or systolic blood pressure.(98) Others have reported similar findings in this highlight the inconsistencies in present trial data.

Although not directly comparable, it is noteworthy that a significant reduction in blood pressure was observed in 281 patients with diabetes and albuminuria after randomisation to the synthetic calcitriol analogue paricalcitol in the Selective Vitamin D Receptor Activator for Albuminuria Lowering Study.(99) The vast majority of these patients were hypertensive at entry.(99). Increased arterial stiffness as determined by pulse wave velocity has been associated with low 25(OH)D concentrations.(100) In a recent placebo controlled RCT in 120 patients with CKD3/4 and vitamin D insufficiency (25(OH)D < 50nmol/L), cholecalciferol 300,000 IU at baseline and 8 weeks significantly decreased pulse wave velocity (1.24 (2.16 – 0.74) m/s, p<0.001) compared with placebo after 16 weeks (secondary outcome). Similarly, Levin et al noted that in a 3-arm double blind RCT of 119 patients with CKD3b/4, calcidiol (25(OH)D) 5000IU 3x/week and calcitriol 0.5ug 3x/week both decreased PWV when compared to placebo.(101) Given the strong association of PWV with cardiovascular events and mortality, and also as a surrogate for arterial calcification, these data are of considerable importance.

Carotid intima-media thickness (CMT) is a marker of atherosclerosis and has been linked to vitamin D deficiency.(81) Unfortunately, few randomized trials have assessed the effects of vitamin D supplementation on CMT in the general population. In one trial that included 80 patients randomized to receive either 50,000 IU vitamin D/week or placebo for 16 weeks, the investigators found no significant different in CMT. However, pro-atherogenic inflammatory markers such as IL-6 levels were reduced.(67) Further, cholecalciferol may also have an impact on endothelial function. Kumar et al showed a significant increase in brachial arterial flow-mediated dilatation (FMD) with cholecalciferol but not placebo.(102) This is consistent with findings from the Paricalcitol and Endothelial Function in Chronic Kidney Disease (PENNY) trial, where paricalcitol 2ug/day increased FMD when compared to placebo.(103) In contrast, other RCTs have shown conflicting results. Kendrick et al reported no effect of cholecalciferol 2000IU per day on FMD over 6 months, possibly due to the lower dose of cholecalciferol used.(104) Chandler, et al. showed that vitamin D supplementation (at 1000, 2000 or 4000 IU/d orally compared to placebo for 3 months) did not change inflammatory markers in African Americans.(105)

The effect of Vitamin D supplementation on lipid profile and dyslipidemia is unclear, with conflicting results from several randomized trials. In one small randomized trial that included 8 patients with metabolic syndrome receiving 50,000 IU vitamin D or placebo weekly for 16 weeks, only triglyceride levels were significantly reduced.(106) However, in a large retrospective cohort study by Ponda et al., raising vitamin D levels from < 20 to ≥30 ng/ml (n=6260) compared with remaining at < 20 ng/ml (n=2332) was associated with an increase in total cholesterol (0.77 mg/dL, 95% CI 0.18–1.36) and high-density lipoprotein (0.42 mg/dL, CI 0.08–0.76) but no significant changes in low-density lipoprotein or triglycerides. In a meta-analysis that assessed the effects of vitamin D supplementation on blood lipids that included a total of 1346 participants, no statistically significant changes were observed for cholesterol, high-density lipoprotein, or triglyceride levels.(107)

These and other data have led to two recent large randomised placebo-controlled trials in the general population. The ViDA (Vitamin D Assessment) trial enrolled 5,108 participants in New Zealand aged 50–84 years.(108) In the treatment arm, participants received an initial dose of 200,000 IU followed a month later by monthly 100,000 IU or placebo for a median of 3.3 years. The investigators found no significant reduction in incident CVD and death in the treatment group compared to placebo (adjusted hazard ratio 1.02; 95% CI, 0.87–1.20). The VITAL trial examined the benefits and risks of cholecalciferol at a dose of 2000 IU per day and marine n-3 fatty acids at a dose of 1g per day (in a 2 × 2 factorial design) for the primary prevention of cardiovascular disease or cancer with median follow-up of 5.3 years.(109) VITAL included a total of 25,871 participants (men aged 50 years of age or older and women aged 55 years of age or older) in a randomized, double-blind, placebo-controlled design. Vitamin D did not lower the incidence of major cardiovascular events (396 in vitamin D group, 409 in placebo group; hazard ratio, 0.97; 95% CI, 0.85–1.12, P=0.69). Results were not modified by baseline serum 25(OH)D levels, with similar results for those above and below a baseline 25(OH)D level of 20 ng/mL. However, it is noteworthy that in both trial populations were mostly vitamin D replete prior to treatment (ViDA 66.25 ± 22.5 nmol/L; VITAL 77 ± 25 nmol/L). While ViDA and VITAL indicate that significant doses of cholecalciferol do not reduce cardiovascular risk in a generally replete population, they did not assess the effect of treating to sufficiency a vitamin D deficient population. Further, ViDA had a median follow-up time of 3.3 years and VITAL 5.3 years. Latency effects cannot be confidently excluded. Other limitations include the monthly dosing schedule in ViDA which is associated with nonphysiological fluctuations in blood levels of vitamin D.(110) There is therefore still a need for trials in vitamin D insufficient populations with a longer follow-up duration.

A summary of recent major randomized trials is provided in table 1.

Table 1:

Recent major RCTs of vitamin D and CVD events or CVD risk factors

Trial name	Population	Intervention	Follow-up	Outcomes and main findings	Year	Ref
The VITAL (Vitamin D and Omega-3) trial	25 871 men 50 years or older and women 55 years or older	Vitamin D3 (cholecalciferol) 2000 IU per day and omega-3 1g per day or placebo	Median 5.3 years	Vitamin D did not lower incidence of major cardiovascular events or incidence of invasive cancer compared to placebo	2018	109
The ViDA (Vitamin D Assessment) trial	5110 community-resident adults, aged 50 to 84 years	Vitamin D3 (200 000IU), followed a month later by monthly 100 000 IU or placebo	Median 3.3 years	Monthly high-dose vitamin D supplementation did not prevent incident CVD and death compared to placebo	2017	108
The effect of vitamin D statues on endothelial function trial	114 post-menopausal women with 25D concentrations between 10 and 60ng/mL	Vitamin D3 2500 IU or placebo	4 months	No significant differences between groups in changes in brachial artery flow-mediated vasodilation (FMD), pulse wave velocity (PWV), augmentation index (AI) or CRP.	2012	67
The effect of vitamin D repletion on small low density lipoprotein (LDL) particle number in subjects at elevated cardiovascular risk trial	151 adults with 25D concentrations < 20 ng/mL	Vitamin D3 50 000 IU weekly or placebo	8 weeks	Vitamin D repletion did not change small low-density lipoprotein (LDL) levels, total cholesterol, LDL cholesterol, high-density lipoprotein (HDL) or triglyceride levels compared to placebo	2012	144
Calcium/vitamin D supplementation and coronary artery calcification in the Women’s Health Initiative (WHI) trial	754 women aged 50 to 59	Calcium carbonate (1000mg elemental calcium) + vitamin D3 400 IU daily or placebo nested within WHI trial of estrogen among women who underwent hysterectomy	7 years	Treatment with vitamin D3 and calcium did not alter coronary artery calcification (CAC) measurements as assessed by cardiac CT compared to placebo	2010	145
Vitamin D Effects in Overweight Patients (SMART) trial	Healthy overweight subjects (n=200) with mean 25D concentration of 30 nmol/L	Vitamin D preparation (Vigantol oil, Merck) 83μg per day or placebo	12 months	Weight loss was not affected by vitamin D supplementation. Vitamin D group had lower PTH, triglyceride and TNF-α levels compared to placebo.	2009	147
Calcium/vitamin D supplementation and cardiovascular events in the Women’s Health Initiative (WHI) trial	36,282 women aged 50 to 79	Calcium carbonate (1000mg elemental calcium) + vitamin D3 400 IU daily or placebo	7 years	Vitamin D3 and calcium supplementation did not increase or decrease coronary or cerebrovascular risk in healthy postmenopausal women	2007	97
Vitamin D supplementation improves cytokine profiles in patients with congestive heart failure	93 patients with congestive heart failure (CHF), New York Heart Association class ≥ 2	Vitamin D3 50μg daily + 500mg calcium or placebo + 500mg calcium	9 months	Pro-inflammatory cytokine TNF-α increased in the placebo group but remained constant in the vitamin D3 treatment group. There was no change in survival rate between the two groups.	2006	146
Effect of four monthly oral vitamin D3 (cholecalciferol) supplementation on fractures and mortality in men and women living in the community	2686 adults aged 65–85 years	Vitamin D3 100 000 IU or placebo every four months	5 years	There was no significant difference in all-cause mortality or cardiovascular disease between the two groups.	2003	148
Vitamin D and the health of blood vessels in kidney disease trial	119 patients eGFR 15–45 ml/min/1.73m2	Calcifediol 5000 IU, calcitriol 0.5μg or placebo thrice weekly	6 months	PWV decreased in the calcifediol group, remained unchanged in the calcitriol group and increased in the placebo group.	2017	101
The OPERA trial (Effect of Paricalcitol on Left Ventricular Mass and Function in CKD) trial	60 patients with stage 3–5 CKD	Paricalcitol 1μg per day or placebo	52 weeks	Paricalcitol significantly reduced PTH, alkaline phosphatase levels, and cardiovascular-related hospitalizations compared to placebo. There was no change in LV mass index as determined by cardiac MRI	2014	134
The PRIMO (Paricalcitol Capsule Benefits in Renal Failure– Induced Cardiac Morbidity) study	227 CKD patients with mild to moderate LVH and preserved EF	Paricalcitol 2μg per day or placebo	48 weeks	Paricalcitol reduced PTH levels within 4 weeks. At 48 weeks, there was no difference in left ventricular mass index compared to placebo as determined by cardiac MRI	2012	133
Selective vitamin D receptor activation with paricalcitol for reduction of albuminuria in patients with type 2 diabetes study	281 type-2 diabetic patients receiving RAAS inhibitor	Paricalcitol 1μg or 2μg per day, or placebo	24 weeks	Addition of 2μg per day to RAAS inhibition lowered residual albuminuria in patients with diabetic nephropathy	2010	99

LESSONS FROM KIDNEY DISEASE

CVD remains the leading cause of death among patients with CKD. End-stage renal disease (ESRD) patients on dialysis have a 10- to 30-fold higher CVD mortality compared with the general population.(111) CKD patients experience both accelerated atherosclerotic and non-atherosclerotic CVD, involving arterial calcification and left ventricular (LV) hypertrophy with the consequence of increased incidence of arrhythmia, congestive heart failure and sudden cardiac death.(112) CKD aggravates pre-existing traditional risk factors for atherosclerosis, such as hypertension and dyslipidemia. However, non-traditional or novel risk factors such as uremia, pro-inflammatory cytokines, mineral dysregulation and volume shifts on dialysis can all contribute to the development of accelerated atherosclerosis.(113,114) Perturbed vitamin D metabolism is a hallmark of kidney failure and CKD therefore represents a valuable model for vitamin D deficiency. Untreated kidney failure is associated with the classical skeletal features of vitamin D deficiency.

Plasma 25(OH)D concentrations begin to decline when the glomerular filtration rate (GFR) falls below 50 ml/min/1.73m².(115) There are multiple contributing factors to the vitamin D deficiency observed in CKD, including urinary losses in patients with proteinuric kidney disease,(116) impaired hepatic 25-hydroxylation in the context of uremia,(117) reduced dietary intake, and increased catabolism through induction of the catabolic 24-hydroxylase(118) in addition to the known risk factors for deficiency. Further, elevated concentrations of FGF-23 induce the vitamin D catabolic pathway in the kidney and simultaneously inhibit 1α-hydroxylation. Once patients reach ESRD, only a minority are vitamin D sufficient. In a cohort of 6,518 German dialysis patients, 76% were vitamin D deficient,(119) with smaller studies reporting deficiency in 69 – 87%.(120–124)

Approximately 80% of all vitamin D metabolites are carried in blood bound to vitamin D binding protein (VDBP). A 58kDa protein, VDBP is a negative acute phase reactant with a short half-life (1–2 days). In acute illness, VDBP concentrations (and consequently total vitamin D concentrations) reduce. In patients with proteinuric kidney disease, urinary losses of VDBP and its cargo of vitamin D metabolites occur.(116) It is also readily lost through plasma exchange,(124) and these factors may contribute to its deficiency in patients with CKD. Given these variations in VDBP (and their impact on measured total vitamin D), some have proposed determining ‘free’ or bioavailable vitamin D. Indeed, bioavailable vitamin D appears to correlate more closely with markers of mineral metabolism in CKD than does total 25(OH)D.(125) The discovery of genetic polymorphisms in VDBPs and racial differences have added further dimensions of complexity to this field(126).

Challenges with conventional use of 1α-hydroxylated compounds

Given the high prevalence of vitamin D deficiency in CKD populations and the association with adverse health outcomes of particular relevance to CKD, existing CKD treatment guidelines recommend measuring 25(OH)D and replenishing vitamin D with cholecalciferol, aiming for a target threshold of 75 nmol/L (30 ng/mL) or above.(20) However, this is not consistently done since supportive evidence is limited (The Kidney Disease Improving Global Outcomes - KDIGO treatment guidelines grades this as a 2C recommendation), vitamin D measurement is costly, and it is unclear whether the associations observed with deficiency apply in the context of concomitant treatment with 1α-hydroxylated vitamin D; osteomalacia and rickets are both effectively treated with 1,25(OH)₂D,(127) and it is not currently known whether the same applies to other sequelae of vitamin D deficiency in patients with CKD.

Although PTH suppression is readily achieved through administration of 1α-hydroxylated vitamin D, the effect this might have on tissue concentrations of 1,25(OH)₂D is unknown. The existing treatment paradigm of reliance on 1α-hydroxylated vitamin D compounds presents several concerns: 1) systemic administration increases plasma calcium and phosphate; 2) dose is limited by hypercalcemia; 3) it may potentiate vascular calcification and 4) potent induction of the vitamin D catabolic pathway occurs, paradoxically potentiating vitamin D deficiency by catabolism of 1,25(OH)₂D and 25(OH)D at tissue level. In contrast, the administration of inactive cholecalciferol could circumvent all of these limitations. The systemic administration of cholecalciferol results in high tissue concentrations of 1,25(OH)₂D without the same risk of causing hypercalcemia seen with 1α-hydroxylated VDRAs.(128,129) Furthermore, cholecalciferol use results in significant reductions in PTH, and either reduced or unchanged plasma calcium and phosphate concentrations.(122,130) Cholecalciferol also results in reductions in bone alkaline phosphatase (ALP), and improves the achievement of treatment targets for calcium, phosphate and PTH concentrations.(122)

The fact that the VDR and CYP27B1 are ubiquitously expressed mandates a reappraisal of the current practice of administering mostly 1α-hydroxylated compounds to patients with ESRD, since this strategy was predicated on the now disproven belief that 1α-hydroxylation was exclusive to the kidney. The ubiquitous use of 1α-hydroxylated compounds (alfacalcidol, calcitriol, paricalcitol etc.) in CKD presents the greatest challenge to interpretation of evidence from studies in the general population to the CKD setting.

Therapeutic targeting of CVD in CKD with vitamin D

In a retrospective observational study, Lishmanov et al compared patients with CKD3/4 and vitamin D < 75 nmol/L that were prescribed ergocalciferol replacement (n=90) to those that received no replacement (n=30). Patients receiving ergocalciferol had a significantly lower risk of cardiovascular events over a mean follow-up of 27.2 months (HR 0.37, 95% CI 0.14 – 1.0).(131) While observational studies such as this support therapeutic supplementation of vitamin D in CKD patients for cardiovascular benefit, unfortunately interventional studies have demonstrated conflicting results. It should be noted that randomized controlled trials that specifically examine atherosclerosis as the primary prespecified outcome in CKD patients are severely lacking.

In the recent J-DAVID (Japan Dialysis Active Vitamin D) trial, 976 hemodialysis patients with serum PTH levels less than or equal to 180pg/ml were recruited.(132) The patients were randomized to receive either 0.5μg oral alfacalcidiol per day or no vitamin D receptor activator with median follow-up of 4 years. The investigators found no significant difference in composite outcome of cardiovascular events in the intervention group (21.1%) compared to the control group (17.9%; absolute difference 3.25% [95% CI, −1.75% to 8.24%; HR 1.25 [95% CI, 0.94–1.67], p=0.13).

The PRIMO (Paricalcitol Capsule Benefits in Renal Failure – Induced Cardiac Morbdity) study recruited 227 patients in a multinational double-blinded placebo-controlled trial designed to study the effects of paricalcitol on left ventricular mass index (LVMI) using cardiac magnetic resonance (CMR) imaging (133). G3 and G4 CKD patients with mild-to-moderate left ventricular hypertrophy (LVH) and preserved LV ejection fraction at baseline were included and followed over a period of 48 weeks. The study found that paricalcitol did not alter LVMI (0.34 g/m2.7, 95% CI −0.14–0.83 g/m2) compared to placebo (−0.07, g/m2.7, 95% CI −0.55–0.42 g/m2). In the paricalcitol arm, PTH levels were reduced to 30% of baseline, markedly increased concentrations of serum calcium and phosphate and a lower number of cardiovascular hospitalizations. There was no change in pre-specified measures of diastolic function, or SBP. The OPERA trial (Effect of Paricalcitol on Left Ventricular Mass and Function in CKD) was a prospective, double-blinded, randomized, placebo-controlled trial to determine whether oral activated vitamin D reduced LV mass as examined by echocardiography (134). Patients were randomly assigned to receive either oral paricalcitol (n=30) or matching placebo (n=30) for 52 weeks. Similar to the PRIMO study, the study did not find a difference in LV mass between the two arms.

Despite several animal models of LVH that have shown promising therapeutic effects of vitamin D, including the spontaneously hypertensive rat model(135), the Dahl salt sensitive rat(136) and the pressure overload rat(137), clinical trials have not shown benefit. A number of possibilities exists that could explain these findings, such as treatment duration that may have been too short to detect a difference, the degree of LVH was too far advanced, or high FGF-23 levels that can directly induce LVH development(138) and mask beneficial effects of vitamin D. One should also entertain the possibility that the primary hypothesis that vitamin D can reduce cardiovascular disease in CKD is in fact null in humans.

There is evidence linking vitamin D deficiency with SCD in hemodialysis patients.(139) Vitamin D treatment of hemodialysis patients has been associated with a reduced QTc interval, prolongation of which is a significant risk factor for SCD(140). In one large cross-sectional study that recruited 3299 Caucasian patients who were routinely referred to coronary angiography, low 25(OH)D and 1,25(OH)₂D concentrations were associated with SCD(77).

HOW MUCH VITAMIN D IS ENOUGH?

Vitamin D status is determined by measuring 25(OH)D concentrations. The 2011 report by the IOM suggests that 25(OH)D levels of 16 ng/ml (40nmol/liter) meet the needs of approximately half the population (median population requirement) and levels of 20ng/ml (50 nmol/liter) meet the needs of at least 97.5% of the population. These findings consider the observation that lower concentrations were associated with rickets and osteomalacia (21) and that thresholds are reached for intestinal absorption of calcium and for bone mineral density at these 25(OH)D concentrations, There remains controversy as to the definitions of vitamin D deficiency and varying definitions have caused confusion. For example, the UK Department of Health and Scientific Advisory Committee on Nutrition (SACN) define vitamin D deficiency as < 25 nmol/l (10ng/ml). Others have argued that the deficiency threshold should be significantly higher, at 50 nmol/L (20 ng/dL).(10) There is additional controversy regarding the upper limit of normal or tolerable upper intake levels; different cut-offs have been proposed ranging from 50 ng/ml to 150ng/ml. Excess vitamin D, usually at 25(OH)D concentrations > 150ng/ml can cause vitamin D toxicity that results in hypercalcemia, renal failure, neuropsychiatric and gastrointestinal symptoms as well as cardiovascular manifestations such as hypertension, arrhythmia and heart block. In extreme cases, vitamin D toxicity can cause coma and death.(141)

Although the thresholds for defining sufficiency have historically been based on skeletal health, there is increasing recognition that non-skeletal disease outcomes should also be taken into account. For example, while levels > 10 ng/ml are optimal to prevent rickets and osteomalacia, levels > 30 ng/ml may be needed to prevent both secondary hyperparathyroidism and osteoporosis. However, the utility of PTH concentrations for identifying the optimal level of vitamin D remains controversial and the relationship between 25(OH)D and PTH is inconsistent.(15) Additionally, recommendations should take into account the fact that vitamin D concentrations are influenced by sunlight exposure, increasing latitude, skin pigmentation, adiposity, low physical activity, poor diet, diabetes and CKD.(142)

The Recommended Dietary Allowance (RDA) provided by the IOM for vitamin D is 600 IU per day for individuals 1 to 70 years of age and 800 IU/d for those above 70 years of age. These recommendations are pooled from 32 study averages of serum 25(OH)D concentrations. Per the IOM, this RDA is expected to achieve a 25(OH)D level of 50 nmol/L or more in 97.5% of healthy individuals.(18) A common misconception is that the RDA defines a threshold or “cut point” of which virtually the entire population must have a serum 25(OH)D concentration above 20ng/ml to achieve adequate bone health.(15) The reality is that most of the population (about 97.5%) has a requirement of 20ng/ml or less and approximately half of the population has a requirement of 16ng/ml or less. (15)

A number of current trials are seeking to reduce the uncertainty over the benefits of vitamin D supplementation. VITAL will soon report on additional health outcomes, including heart failure, cognition, diabetes, depression, and autoimmune diseases. The D-Health trial is assessing the effect of 60,000IU cholecalciferol monthly on all cause and cancer mortality in 21,315 Australian adults.(143) The Finnish Vitamin D Trial (FIND) is assessing the effects of up to 3,600IU daily of cholecalciferol in reducing cancer and cardiovascular risk in 2495 participants in men aged 60 years or older and women 65 years or older ( NCT01463813). DO-HEALTH is testing vitamin D₃ 2000 IU daily for bone health, fall reduction, physical performance, and other outcomes among >2000 European adults. One large randomised trial of cholecalciferol versus standard care in dialysis is currently underway in the United Kingdom (The SIMPLIFIED Trial, EudraCT 2015–005003-88). These and other trials (Table 2) should generate evidence for the general and CKD population of the efficacy and effectiveness of vitamin D in reducing cardiovascular and cancer risk.

Table 2:

Ongoing Vitamin D clinical trials

Trial name	Population	Intervention	Main outcomes	Results expected	Country
VIDAL (Vitamin D and Longevity trial)	20,000 adults, 65–85 years old	100,000 IU D3 monthly, placebo or open control	Mortality, morbidity (infections, doctor’s visits, cancer) and vitamin D levels	2020	United Kingdom
SIMPLIFIED (Survival Improvement with Cholecalciferol in Patients on Dialysis)	4,200 adults, aged > 18 years	Cholecalciferol 60,000 IU fortnightly	Mortality, health-related quality of life, hospital admission, cardiovascular events, cancer incidence	2023	United Kingdom
TARGET-D (Trial Evaluating Vitamin D Normalization on Major Adverse Cardiovascular-Related Events Among Myocardial Infarction Patients)	890 adults, aged > 18 years	Target 25D levels > 40ng/mL with cholecalciferol or standard of care (no intervention)	Mortality, myocardial infarction, heart failure hospitalization and cerebral vascular accident	2021	United States
FIND (Finnish Vitamin D Trial)	2495 adults, men > 60 years and women > 65 years	3200 D3 daily, 1600 D3 daily or placebo	Cardiovascular disease, cancer and diabetes mellitus	2019	Finland
Effect of Vitamin D Supplementation in Patients with Heart Failure and Vitamin D Deficiency	60 adults, aged 45 to 85 years	5 000IU vitamin D D vs placebo	Myocardial structure and function by MRI, myocardial infarction	2019	Mexico
Effect of Vitamin D on Ventricular Remodeling in Patients with Acute Myocardial Infarction (VITDAMI)	144 adults, aged 40 to 85 years	15,690 IU Calcifediol vs placebo	Change in cardiac remodeling by MRI, echo parameters	2019	Spain
Magnesium and Vitamin D Supplementation and Cardiometabolic Outcomes	123 adults, aged 30 to 70 years	1000U vitamin D and 360 magnesium vs 1000U vitamin D vs placebo	PTH levels, inflammatory cytokines, lipid profile, blood pressure, serum osteocalcin	2020	United States
Vitamin D Prophylaxis in the Prevention of Hypertensive Disorders of Pregnancy	412 adults, aged > 18 years	3000 IU vitamin D daily or 4000 IU daily with and without prenatal vitamins	Blood pressure, new onset proteinuria, thrombocytopenia, impaired liver function testing, renal insufficiency	2019	United States
DO-HEALTH (Vitamin D3-Omega3-Home Exercise – Healthy Ageing and Longevity Trial)	2157 adults, aged 70 years and older	1) 2000 IU vitamin D compared to placebo; 2) 1g omega-3 fatty acid compared to placebo; 3) home exercise of 30 minutes 3 times a week compared to control exercise program 30 minutes 3 times a week	Incident non-vertebral fractures, functional decline (measured by short physical performance test battery), blood pressure, cognitive decline (by montreal cognitive assessment), immunity (by rate of any infections)	2019	Switzerland

Conclusion

Clear scientific evidence supports a key role for vitamin D in skeletal health. However, there is inconclusive evidence to support benefits of vitamin D supplementation for extraskeletal health outcomes. Furthermore, existing evidence suggests that most of the population is meeting their requirements, at suggested RDAs by IOM and at the 25(OH)D level of at least 20ng/ml (50 nmol/liter) even under conditions of low sun exposure.(21) Additional trials assessing whether vitamin D supplementation in populations with lower serum 25(OH)D levels, with treatment and follow-up over longer durations, and testing higher dosages of vitamin D, will confer CVD benefits are needed. Also, additional data on cardiovascular outcomes such as heart failure will be of great importance.

In high risk groups such as the CKD population, however, vitamin D deficiency is prevalent. The alternative strategy of administering 1α-hydroxylated compounds for treatment of secondary hyperparathyroidism and renal osteodystrophy has prevailed over recent decades and continues to do so in this population, but the rationale for the preferential use of these drugs is not strong, nor supported by robust evidence from adequately powered randomized trials. For patients not currently enrolled in trials, treating established vitamin D deficiency with cholecalciferol appears to be a rational and reasonable way forward while further evidence accumulates.