The role of vitamin D in the pathogenesis and treatment of diabetes mellitus: a narrative review
Diabetes mellitus, a metabolic disorder associated with chronic complications, is traditionally classified into two main subtypes. Type 1 diabetes mellitus (T1DM) results from gradual pancreatic islet β cell autoimmune destruction, extending over months or years. Type 2 diabetes mellitus (T2DM) is a heterogeneous disorder, with both insulin resistance and impairment in insulin secretion contributing to its pathogenesis. Vitamin D is a fat-soluble vitamin with an established role in calcium metabolism. Recently, several studies have provided evidence suggesting a role for it in various non-skeletal metabolic conditions, including both types of diabetes mellitus. Preclinical studies of vitamin D action on insulin secretion, insulin action, inflammatory processes, and immune regulation, along with evidence of an increase of hypovitaminosis D worldwide, have prompted several epidemiological, observational, and supplementation clinical studies investigating a potential biological interaction between hypovitaminosis D and diabetes. This narrative review aims to summarize current knowledge on the effect of vitamin D on T1DM and T2DM pathogenesis, prevention, and treatment, as well as on micro- and macrovascular complications of the disease. Furthermore, on the basis of current existing evidence, we aim to highlight areas for potential future research.
KeywordsVitamin D Cholecalciferol Type 1 diabetes mellitus Type 2 diabetes mellitus Diabetes complications
Type 1 diabetes mellitus (T1DM) is a chronic metabolic disorder resulting from gradual pancreatic islet β cell destruction because of a prolonged autoimmune process, extending over months or years . Type 2 diabetes mellitus (T2DM), the most prevalent form of diabetes, is a heterogeneous disorder commonly associated with insulin resistance in the presence of an impairment of compensatory insulin secretion .
Vitamin D is a fat-soluble vitamin synthesized in the skin from 7-dehydrocholesterol in a non-enzymatic two-step process. Vitamin D can act as a hormone mainly through a nuclear vitamin D receptor (VDR). Although vitamin D has primarily been studied with regard to its role in bone metabolism and calcium, phosphorus, and magnesium homeostasis, several other biological actions have been investigated, resulting in a plethora of evidence, suggesting a possible role of hypovitaminosis D in the pathogenesis of various non-skeletal metabolic conditions, including diabetes mellitus .
The hypothesis of a possible cause-and-effect relationship between hypovitaminosis D and diabetes was supported by the detection of VDR in the pancreatic β cells , as well as in other tissues involved in glucose homeostasis, such as skeletal muscle and adipose tissue [4, 5]. Specific VDR polymorphisms have been associated with altered insulin secretion and sensitivity [6, 7] implicating vitamin D in the pathogenesis of both types of diabetes. Some decades ago, ecological studies on diabetes (mainly TIDM) prevalence, along with reports on the “hypovitaminosis D epidemic” worldwide [8, 9], raised the hypothesis of a possible relationship between these two pathophysiological entities.
Since then, evidence from several epidemiological and observational studies has pointed to a potential biological interaction between hypovitaminosis D and the pathogenesis, prevention, glycemic control, and micro- and macrovascular complications of both T1DM and T2DM. However, epidemiological and observational studies cannot prove causality. Hence, randomized clinical trials have been conducted to establish or disprove the beneficial effect of vitamin D supplementation on the prevention and/or the clinical course of the disease. This narrative review aims to summarize current knowledge on the effect of vitamin D on T1DM and T2DM pathogenesis, prevention, and treatment, as well as on micro- and macrovascular diabetes complications.
Vitamin D and type 1 diabetes mellitus
The role of vitamin D in the predisposition, pathogenesis, prevention, and glycemic control of T1DM has been thoroughly investigated. Most studies agree on a potential beneficial effect of vitamin D for this specific population, but highlight the variety and the complexity of this beneficial effect, depending on the genetic background, the stage of the disease, and the supplementation regimen.
Predisposition and pathogenesis
Preliminary evidence initially implicated specific VDR gene polymorphisms in variations in insulin secretory capacity . Since then, VDR gene polymorphisms (Bsm-I, Apa-I, Foq-I, and Taq-I) have been demonstrated to be involved in either an aggravating or a prophylactic role in the pathogenesis of T1DM in genetically susceptible individuals [11, 12, 13, 14, 15, 16, 17]. However, as genetic predisposition to the disease is a highly complex process, it is more plausible that a cluster of haplotypes contributes to disease susceptibility than isolated VDR gene polymorphisms alone.
A similar association has been proposed for specific vitamin D binding protein (VDBP) polymorphisms and genetic susceptibility to T1DM. Although specific VDBP genotypes have been linked to islet autoimmunity [18, 19] and decreased concentrations of VDBP have been reported in both T1DM patients and pregnant women whose offspring developed T1DM in childhood [20, 21], not all studies are in agreement [20, 22].
The many conflicting results point to the fact that a definite VDBP role in disease pathogenesis remains to be elucidated. In addition, 25-hydroxylase (CYP2R1) and 1α-hydroxylase (CYP27B1) gene polymorphisms (polymorphisms of the enzymes that convert inactive vitamin D to its active form of 1,25-dihydroxyvitamin-D3) have been extensively investigated for their possible association with increased susceptibility to T1DM [23, 24, 25, 26, 27]. Diverse results from different populations highlight the complexity of the genetic traits and underline the contribution of combined haplotypes, along with ethnic differences and environmental factors, to the genetic predisposition to T1DM. More studies on vitamin D-related gene polymorphisms (VDR, VDBP, CYP2R1, and CYP27B1) and their possible impact on T1DM susceptibility are needed until these polymorphisms can be included in the non-HLA susceptibility genetic markers  that have, to date, been confirmed as contributing to the development of the disease.
A potential protective effect of vitamin D against the development of T1DM may be related to its diverse immunomodulatory properties . These properties seem to contribute to the suppression of chronic inflammation (insulitis) in islets of Langerhans involving modification of CD8+ and CD4+ T cell, B lymphocytes, and macrophage infiltration . Animal studies have demonstrated that administration of high doses of calcitriol (1,25(OH)2D3) in non-obese diabetic (NOD) mice have resulted in reversal of chronic insulitis and diabetes occurrence  and suggested a delay in disease progression even when vitamin D analogs were administered after the initiation of the autoimmune process [31, 32]. However, these results have not been confirmed by human studies.
Prevention of T1DM
The seasonal variation of T1DM onset and the higher incidence that has been observed in certain areas at a high geographic latitude [33, 34, 35] have stimulated the investigation into the contribution of vitamin D status to T1DM risk.
Available meta-analyses of previous observational studies have indicated that hypovitaminosis D is highly prevalent in patients with T1DM compared with their age-matched controls  and is associated with an early onset of the disease . On the other hand, although results from most available studies in different ethnic and age groups suggest a possible causal association between hypovitaminosis D and T1DM prevalence [36, 37, 38, 39, 40, 41, 42, 43], a Danish study failed to establish any association between 25(OH)D concentrations and T1DM incidence in children and adolescents .
It has been suggested that vitamin D intake during early life decreases future risk of T1DM . A meta-analysis of eight observational studies comparing vitamin D supplementation with non-supplementation during early life confirmed this hypothesis regarding T1DM, resulting in an estimated pooled odds ratio of 0.71 (95% CI, 0.51–0.98) . However, the importance of this meta-analysis is limited by the inclusion of heterogenous, mostly case-control studies based largely on analysis of questionnaires.
The strongest existing evidence has been provided by a Finnish supplementation study where the relative risk of developing T1DM by the age of 30 among children who were regularly supplemented with vitamin D soon after birth decreased significantly compared with children who had received no supplementation . Notably, the highest protection against T1DM was observed in children that were given at least 2000 IU/day of vitamin D3 , highlighting the importance of early initiation, long duration, and sufficient vitamin D supplementation.
Maternal vitamin D status and intake during pregnancy, maternal VDBP status, and VDBP and VDR gene polymorphisms have also been investigated for their potential contribution to T1DM risk in offspring [21, 39, 45, 46, 47, 48, 49]. However, while the existing evidence does not support a causative role of maternal vitamin D deficiency in T1DM development in offspring, the available data suggest that specific maternal VDR SNPs in the presence of maternal hypovitaminosis D may influence the in utero environment and fetal imprinting towards an early programming of T1DM in the fetus [48, 49]. Considering that most of the existing knowledge comes from observational studies of cross-sectional design, the need for interventional randomized cohort studies with adequate and multiple dosing schedules and long duration of vitamin D supplementation is necessary to fully determine the role of early vitamin D supplementation in T1DM prevention. The Environmental Determinants of Diabetes in the Young (TEDDY) is such a prospective study that examines the roles of dietary and other environmental exposures in the development of islet autoimmunity (IA) and T1D among children carrying high-risk HLA-DR-DQ genotypes .
Hypovitaminosis D is more prevalent in T1DM patients with poor glycemic control, while better glycemic control has been consistently related to higher 25(OH)D concentrations [51, 52]. These observations have been substantiated by interventional studies where vitamin D supplementation has been found to improve glycated hemoglobin (HbA1c) in T1DM patients [53, 54]. However, most studies performed indicate moderate effects at best, while the use of analogs (alfacacidol) in specific subgroups (LADA patients) seems more promising . Based on the above, screening for vitamin D deficiency, especially in the case of poor glycemic control, has been suggested in T1DM, while supplementation studies should target specific patient sub-groups that are likely to benefit the most from similar interventions.
Vitamin D and type 2 diabetes mellitus
Over the last few decades, a potential role of hypovitaminosis D in the pathogenesis of prediabetes and T2DM has been suggested based on evidence from animal and human studies.
Chronic inflammation and defects in insulin action and secretion are the main pathogenetic pathways leading to T2DM, and vitamin D has been suggested as having an effect on each of them.
T2DM is considered a state of low-grade chronic inflammation and is characterized by increased concentrations of acute phase proteins and several inflammation markers . These inflammatory mediators, along with adipose tissue inflammation and secretion of adipocytokines, may result in insulin resistance  and β cell dysfunction . Mediating both innate and adaptive immunity, vitamin D can prevent production of inflammatory cytokines and reduce the chronic low-grade inflammation observed in T2DM . Moreover, the presence of VDRs in skeletal muscle cells suggests a potential role of vitamin D in the major regulatory system of glucose homeostasis in muscle tissue. Vitamin D stimulates the expression of the human insulin receptor gene, hence having an effect on insulin action modulation and insulin responsiveness enhancement . 1,25(OH)2D3 can increase peroxisome proliferator-activated receptor delta (PPAR-δ) gene expression and favorably affect fatty cell accumulation and fatty acid oxidation . 1,25(OH)2D3 can activate transcription of the human insulin gene and thus play an essential role in insulin secretion . However, it must be emphasized that vitamin D is a calciotropic hormone involved in the regulation of calcium influx through cell membranes and concentration of cytosolic free calcium. Therefore, vitamin D effects on insulin secretion and action may be mediated by extracellular and, mainly, intracellular calcium handling [63, 64]. Furthermore, profound hypovitaminosis D results in secondary hyperparathyroidism where increased PTH concentrations promote dephosphorylation of glucose transporter-4 (GLUT-4), leading to diminished insulin-stimulated glucose transport [65, 66, 67].
VDRs have been identified in pancreatic β cells , modifying insulin response to a glucose load in a positive and direct way through stimulation of insulin secretion. Animal studies in mice lacking a functional VDR have shown a decrease in insulin secretion by 60% after glucose loading . Moreover, expression of 1-alpha-hydroxylase by β cells reflects their capacity to locally activate vitamin D. Pancreatic islets can autonomously produce active vitamin D (calcitriol) , this being important since calcitriol promotes β cell biosynthetic capacity and accelerates the conversion of proinsulin to insulin, as has been evidenced by animal studies . Calcitriol may promote insulin secretion in an autocrine or paracrine way .
Risk for developing T2DM
The prevalence of type 2 diabetes mellitus (T2DM) is rising rapidly worldwide. According to the 7th edition of the IDF Diabetes Atlas, 415 million adults currently have diabetes, and this number is expected to rise to 642 million by 2040 . Epidemiological data associating diabetes incidence with high latitude and minimal or reduced sunlight exposure led to the development of the hypothesis of T2DM prevention via vitamin D supplementation.
It has been reported that hypovitaminosis D in rats reduces pancreatic insulin secretion . Animal studies in recent years as well as the vast majority of observational studies and meta-analyses indicate that hypovitaminosis D is associated with a higher risk of developing T2DM [74, 75, 76, 77, 78, 79]. Additionally, long-term maintenance of optimal vitamin D concentrations from early in life has been associated with reduced future risk of developing T2DM , suggesting a new cost-effective strategy for reducing diabetes risk in adulthood and in the elderly in susceptible individuals. However, interventional trials failed to confirm a cause-and-effect relationship and reported no effect of vitamin D supplementation on glycemic markers [81, 82, 83, 84, 85, 86, 87] or T2DM incidence [82, 83] in non-diabetic populations. A recent systematic review of meta-analyses from randomized controlled supplementation studies failed to establish a beneficial effect of vitamin D supplementation on the risk of developing T2DM . However, most available trials have not been designed with the aim of exploring non-skeletal effects of vitamin D supplementation and did not include subjects with hypovitaminosis D. Thus, any potential beneficial effects of vitamin D supplementation in at-risk individuals might have been missed.
Prediabetes, the initial reversible stage of T2DM, represents a “window of opportunity” for designing disease prevention studies. Working in this direction, several trials have been designed to evaluate the potential role of vitamin D supplementation in prediabetic individuals, however, with variable outcomes [89, 90, 91, 92, 93, 94, 95]. No beneficial effect of vitamin D supplementation on insulin resistance indices (HOMA-IR) and 2-h plasma glucose concentration after oral glucose tolerance test (OGTT) in individuals with prediabetes was shown in a recent meta-analysis of ten randomized control trials . However, in a subgroup of participants with baseline serum 25(OH)D < 20 ng/ml, 2-h plasma glucose after OGTT was significantly reduced, indicating that vitamin D supplementation might be beneficial in vitamin D-deficient individuals with prediabetes. On the other hand, longer duration of supplementation did not seem to affect the outcomes, as implied by previous meta-analyses . Conflicting evidence may be due to the lack of available data regarding serum 25(OH)D levels after supplementation. As vitamin D should be considered more as a nutrient hormone rather than a drug, achieving optimal levels might be necessary, as a threshold effect of 25(OH)D level on glycemia has previously been advocated .
Controversial results from supplementation studies may arise from inadequately powered, not appropriately designed studies and not well-defined “at-risk” populations. Μore adequately powered studies investigating vitamin D supplementation in individuals at risk for diabetes are required. The D2d study is such a large multicenter clinical trial that aims to shed light on the complex vitamin D-glucose metabolism relationship, enrolling over 2400 individuals at risk of developing T2DM .
Observational studies showing an inverse association between HbA1c and vitamin D status in T2DM [100, 101, 102] have led to the hypothesis that vitamin D status may play a role in the glycemic control of T2DM patients. Unfortunately, interventional studies resulted in conflicting results due to heterogeneity mostly in dose and duration of vitamin D supplementation and different definitions of vitamin D deficiency. Vitamin D supplementation has been reported to improve serum-fasting plasma glucose and HOMA-IR [103, 104, 105, 106], though no effect was evident in other studies . To clarify this controversy, randomized, double-blinded, placebo control trials have been conducted and meta-analyzed demonstrating a beneficial effect of vitamin D supplementation in severely vitamin D deficient [106, 107, 108] and poorly controlled [109, 110] T2DM patients. In a recent meta-analysis, vitamin D supplementation was associated with reduced HbA1c levels (standardized mean difference (SMD) − 0.25 (− 0.45 to − 0.05)) but had no influence on FBG levels (SMD − 0.14 (− 0.31 to 0.03)) . However, the subgroup analyses suggested that vitamin D supplementation was associated with reduced HbA1c levels (SMD − 0.39 (− 0.67 to − 0.10)) and FBG (SMD − 0.27 (− 0.46 to − 0.07)) in patients with 25(OH)D deficiency at baseline (<50 nmol/l) .
Therefore, current evidence does not support the widespread use of vitamin D supplementation in individuals with T2DM, while current recommendations by the American Diabetes Association do not advocate vitamin D supplementation to improve glycemic control .
Vitamin D and diabetes mellitus complications
A potential association of hypovitaminosis D with long-term micro- and macrovascular diabetes complications has been hypothesized based on the established actions of vitamin D (antiproliferative, immunomodulatory, angiogenic, inhibition of the renin-angiotensin-aldosterone system, and neurotrophic factor expression), which likely interact with the pathogenesis of diabetes complications .
Vitamin D and microvascular complications of diabetes mellitus
Diabetic retinopathy (DR), affecting approximately one third of patients with diabetes, is one of the leading causes of vision loss in working-age adults worldwide . It is commonly accepted that DR pathogenetic pathways, although still under research, lie beyond the duration of diabetes and glycemic control or additional microvascular comorbidities. Established risk factors include family history, age, hypertension, smoking, inflammation, insulin resistance, anemia, dyslipidemia, obesity, and microalbuminuria . DR prevention through timely identification of modifiable risk factors in patients with diabetes mellitus is of paramount importance. Based on the anti-inflammatory and immunosuppressive effects of vitamin D and its potential inhibitory effect on angiogenesis , a cause-and-effect relationship between vitamin D and DR has been hypothesized.
Τhe first indications of an association between vitamin D and DR originated from in vitro and experimental animals studies, where high concentrations of serum 1,25(OH)2D3 were associated with reduced angiogenesis in retinoblastoma models and ischemic retinopathy in mice . In addition, VDR has been identified in the retina of rats  and an inhibitory effect of vitamin D on DR was demonstrated .Human studies investigating the potential interaction between vitamin D status and the presence or severity of DR among T2DM patients have provided sparse and conflicting data [116, 119, 120, 121, 122, 123, 124, 125, 126, 127, 128, 129]. Several limitations are evident in the available observational studies, namely, cross-sectional design, small sample size, ethnic variability, and heterogeneity in the criteria that are not conducive to identification of vitamin D insufficiency, leading to inconclusive results. Two recent meta-analyses of observational studies indicate that hypovitaminosis D among T2DM patients is associated with a significantly increased risk of DR [130, 131].
In addition to vitamin D status, specific VDR gene polymorphisms have been associated with DR in T2DM. A study in a Chinese population showed an association between the VDR gene polymorphism Foq-I and the onset of DR . However, two studies investigating the relationship of Bsm-I polymorphism in the VDR gene with the risk of DR in T2DM patients in different ethnic groups have produced conflicting results [133, 134].
The evidence with regard to T1DM seems more consistent [135, 136]. The prevalence of DR has been found to be significantly higher in T1DM patients with vitamin D deficiency (25 (OH)D < 20 ng/ml), regardless of the duration of diabetes . VDR gene polymorphisms have been linked to DR risk in T1DM, suggesting a possible role of vitamin D in the pathogenesis of DR [137, 138]. A Fok-I single nucleotide polymorphism of the VDR gene has been associated with less severe DR . Similarly, individuals with the TT genotype of the VDR Taq-I polymorphism were less likely to develop DR .
To conclude, based on the existing results of cross-sectional studies concerning DR, hypovitaminosis D could be considered a risk factor for the development and severity of T1DM and T2DM. Prospective and randomized controlled supplementation trials are needed to establish a beneficial effect of vitamin D supplementation in the prevention of DR. Further genetic research on vitamin D-related polymorphisms could provide a useful screening tool for identifying DR-susceptible individuals and candidates who will benefit most from vitamin D supplementation.
Diabetic nephropathy (DN) is characterized by specific structural and functional renal defects associated with T2DM and T1DM . Advanced glycation end products, hemodynamic and endocrine changes, and secretion of growth factors producing reactive oxygen species and inflammatory mediators are the leading pathophysiological causes of DN. As a result, glomerular hyperfiltration, glomerular hypertension, and renal hypertrophy, clinically manifested with proteinuria and hypertension, are present in patients with DN.
Regardless of its basal status, vitamin D has been shown to reduce oxidative stress  and inflammation , maintaining podocyte morphology, preventing epithelial-to-mesenchymal transformation , and suppressing renin gene transcription , thereby potentially reducing the risk of DN. Recent animal studies have suggested that modification in the expression of Klotho and blocking of renin-angiotensin activation may mediate the protective effect of vitamin D on DN .
Familial clustering in the risk for the development of DN has a polygenic component. Among several candidate genes, VDR gene polymorphisms (Apa-I, Bsm-I, Fok-I, and Taq-I) have been studied for their possible protective or aggravating role in susceptibility to DN. Results of relevant studies are contradictory, as were also the results of the two most recent meta-analyses [145, 146], highlighting the need for further well-designed trials to estimate the potential association of VDR gene polymorphisms with DN risk occurrence.
Microalbuminuria is one of the initial clinical manifestations of DN and a well-known indicator of early renal disease . An increased risk of microalbuminuria in the presence of hypovitaminosis D has been identified in T1DM patients . Moreover, vitamin D analogs have been shown to ameliorate proteinuria [149, 150]. Paricalcitol administration further reduced albuminuria among T2DM and T1DM patients who were already on RAAS blockers [151, 152].
However, it could be argued that since (in most studies) patients were already on RAAS blocking medication, further reduction of microalbuminuria was not a result of vitamin D therapy per se, hence the studies could not support a causality between vitamin D analog therapy and urinary albumin excretion reduction in diabetic patients. Further investigation of the use of vitamin D supplementation to prevent and reduce microalbuminuria in patients with diabetes is required. To our knowledge, no evidence linking vitamin D depletion or impaired vitamin D metabolism to early glomerular filtration rate (GFR) loss in patients with diabetes has been documented so far. Therefore, vitamin D replacement for prevention of DN in clinical practice cannot currently be recommended.
Diabetic peripheral neuropathy
Diabetic peripheral neuropathy (DPN) is a chronic disabling complication of diabetes mellitus , affecting about half of the diabetes population. Demyelination and axonal degeneration are established pathogenetic hallmarks of diabetic neuropathy, possibly in addition to microvascular and extracellular nerve pathology . Vitamin D deficiency has been linked to the pathogenesis of DPN, given its role as a neurotrophic factor . Notably, VDR and CYP27B1 expression in different parts of the nervous system  strongly point to a significant modulatory role of vitamin D in neuronal cells. A causal association between hypovitaminosis D and several other neurological diseases, such as multiple sclerosis, schizophrenia, Parkinson’s disease, and dementia, has also been suggested.
Observational studies in T2DM patients suggest that hypovitaminosis D is associated with higher incidence of DPN, as well as the severity of the symptoms caused by DPN. Moreover, recent studies suggest a relationship between the incidence of foot ulcers and hypovitaminosis D . Serum 25(OH)D concentration may be an independent predictor of DPN, as has been demonstrated by studies in different ethnic groups [120, 158, 159]. Results from a recent meta-analysis suggested that hypovitaminosis D is independently associated with increased risk for the development of DPN in patients with T2DM . The clinical question of optimizing vitamin D levels to reduce the risk of developing DPN remains to be elucidated since, to our knowledge, evidence to support such an effect is lacking. On the other hand, a recent observational study has shown a U-shaped association between vitamin D levels and DPN . This study was the first to emphasize the need for vitamin D supplementation monitoring, implying a narrow therapeutic range of optimal vitamin D concentrations for DPN patients.
Clinical symptoms such as chronic pain related to DPN have been associated with hypovitaminosis D. Indeed, oral vitamin D supplementation has been shown to improve neuropathic symptoms by 50% in vitamin D-deficient diabetic patients [162, 163]. Restoration of insulin secretion, reduction of insulin resistance, and reduction in inflammatory response have been proposed as potential mechanisms for the improvement of clinical manifestations of DN following vitamin D supplementation in this specific population [160, 164, 165].
Most of the available evidence on the role of vitamin D in the development of DPN as well as its use as therapeutic approach derives from studies in T2DM patients. Results concerning T1DM are limited . Randomized controlled trials in both T2DM and T1DM are thus needed to make concrete therapeutic recommendations. These studies should address controversial issues, such as the initial and optimal target vitamin D concentration for therapeutic intervention or the best replacement regimens.
Vitamin D and macrovascular complications of diabetes mellitus
The etiology of the increased cardiovascular risk that exists even before disease diagnosis in T2DM patients lies well beyond glycemic control. Epidemiological studies have repeatedly shown the inverse association between vitamin D status and acute myocardial infarction prevalence  and prognosis , risk of stroke , and risk of hospitalization for heart failure . Different pathogenetic mechanisms have been hypothesized, including vitamin D effects on endothelial function, chronic inflammation, renin-angiotensin-aldosterone system regulation, and calcium homeostasis .
Few studies have evaluated the value of serum 25(OH)D levels in predicting cardiovascular disease (CVD) events or the possibility of a protective role of vitamin D supplementation in patients with T2DM. In this direction, cross-sectional , prospective , and cohort [119, 174] studies in T2DM populations have shown an inverse association between vitamin D concentrations and CVD events. Diabetic patients with 25(OH)D < 36 nmol/l manifested approximately 21% higher risk for developing macrovascular disease compared with those with 25(OH)D > 63 nmol/l .
However, prospective studies have failed to demonstrate any benefit from vitamin D supplementation in reducing the incidence of CVD events and diabetes macrovascular complications [175, 176]. The wide heterogeneity of the available results and the relatively small number of studies are not adequate to support vitamin D supplementation in this context.
It is thus evident that subgroups of diabetic populations who would benefit the most from vitamin D supplementation therapy should be identified. Surrogate markers, such as a specific genotype (vitamin D-related gene polymorphisms) or phenotype (increased CVD risk), are needed to identify these subgroups. In fact, a protective or predisposing role for coronary heart disease has been attributed to specific VDR gene polymorphisms . In this regard, these polymorphisms have been proposed as such markers, though the results remain controversial [178, 179, 180, 181]. Carotid intima-media thickness (IMT) and arterial stiffness, both considered early markers of preclinical atherosclerosis, have been used to identify diabetes patients at risk and to monitor intervention effects on the progression to overt CVD. These two surrogate markers could be used in designing interventional studies, since hypovitaminosis D has been independently associated with increased arterial stiffness in prediabetic patients , T2DM patients , and adolescents with T1DM .
Regarding T1DM, most studies provide evidence of an association between 25(OH)D concentration and endothelial function , coronary calcification , and vascular stiffness . However, supplementation has not resulted in any effect on macrovascular events , although available results are scarce and therefore inconclusive.
DM may consist of five different types rather than two, as has recently been suggested by researchers analyzing data from a Swedish cohort . Therefore, the complexity of its pathogenesis makes the contribution of any isolated factor less likely to be proven universally important. In this context, vitamin D contribution to the pathogenesis of T1DM and T2DM has been extensively investigated but not unequivocally proven by supplementation trials. Therefore, more studies implementing better phenotyping or even genotyping of patients and better vitamin D indices (free vitamin D, other calciotropic hormones) are required to identify individuals who would potentially benefit from vitamin D supplementation and to establish an appropriate dosing regimen and length of therapy. Until such results are available, DM patients with risk factors for vitamin D deficiency or metabolic bone disease or with established vitamin D deficiency should be treated with vitamin D supplementation according to the national or international guidelines for vitamin D supplementation in at-risk populations.
Compliance with ethical standards
Conflict of interest
The authors declare that they have no conflict of interest.
- 1.Masharani U, German M (2017) Pancreatic hormones and diabetes mellitus. In: Gardner D, Shoback D (eds) Greenspan’s basic and clinical endocrinology. Lange, McGraw-Hill Education, New York, pp 609–612Google Scholar
- 27.Frederiksen BN, Kroehl M, Fingerlin TE, Steck AK, Rewers M, Norris JM (2013) Association between vitamin D metabolism gene polymorphisms and risk of islet autoimmunity and progression to type 1 diabetes: the diabetes autoimmunity study in the young (DAISY). J Clin Endocrinol Metab 98(11):E1845–E1851PubMedPubMedCentralCrossRefGoogle Scholar
- 31.Casteels K, Waer M, Laureys J et al (1998) Prevention of autoimmune destruction of syngeneic islet grafts in spontaneously diabetic nonobese diabetic mice by a combination of a vitamin D3 analog and cyclosporine. Diabetes 51:1367–1374Google Scholar
- 58.Sung CC, Liao MT, Lu KC, Wu CC (2012, 2012) Role of vitamin D in insulin resistance. J Biomed Biotechnol:634195Google Scholar
- 67.Karras SN, Anagnostis P, Antonopoulou V et al (2017) The combined effect of vitamin D and parathyroid hormone concentrations on glucose homeostasis in older patients with prediabetes: a cross-sectional study. Diab Vasc Dis Res 1:1479164117738443Google Scholar
- 68.Johnson JA, Grande JP, Roche PC, Kumar R (1994) Immunohistochemical localization of the 1,25(OH)2D3 receptor and calbindin D28k in human and rat pancreas. Am J Phys 267:356–360Google Scholar
- 72.International Diabetes Federation (2015) IDF Diabetes Atlas, 7th edn. International Diabetes Federation, BrusselsGoogle Scholar
- 80.Wu F, Juonala M, Pitkänen N et al (2017) Both youth and long-term vitamin D status is associated with risk of type 2 diabetes mellitus in adulthood: a cohort study. Ann Med 7:1–9Google Scholar
- 85.Moreira-Lucas TS, Duncan AM, Rabasa-Lhoret R et al (2017) Effect of vitamin D supplementation on oral glucose tolerance in individuals with low vitamin D status and increased risk for developing type 2 diabetes (EVIDENCE): a double-blind, randomized, placebo-controlled clinical trial. Diabetes Obes Metab 19(1):133–141PubMedCrossRefPubMedCentralGoogle Scholar
- 90.Mitri J, Dawson-Hughes B, Hu FB et al (2011) Effects of vitamin D and calcium supplementation on pancreatic beta cell function, insulin sensitivity, and glycemia in adults at high risk of diabetes: the calcium and vitamin D for diabetes mellitus (CaDDM) randomized controlled trial. Am J Clin Nutr 94:486–494PubMedPubMedCentralCrossRefGoogle Scholar
- 107.Gulseth HL, Wium C, Angel K, Eriksen EF, Birkeland KI (2017) Effects of vitamin D supplementation on insulin sensitivity and insulin secretion in subjects with type 2 diabetes and vitamin D deficiency: a randomized controlled trial. Diabetes Care 40(7):872–878PubMedCrossRefPubMedCentralGoogle Scholar
- 112.Association American Diabetes (2018) Pharmacologic approaches to glycemic treatment: standards of medical care in diabetes. Diabetes Care 41(Suppl 1):S73Google Scholar
- 148.de Boer IH, Sachs MC, Cleary PA et al (2012) Diabetes Control and Complication Trial/Epidemiology of Diabetes Interventions and Complications Study Research Group. Circulating vitamin D metabolites and kidney disease in type 1 diabetes. J Clin Endocrinol Metab 97(12):4780–4788PubMedPubMedCentralCrossRefGoogle Scholar
- 163.Lee P, Chen R (2008) Vitamin D as an analgesic for patients with type 2 diabetes and neuropathic pain. Arch Intern Med 168:171–172Google Scholar
- 170.Costanzo S, De Curtis A, Di Castelnuovo A et al (2018) Serum vitamin D deficiency and risk of hospitalization for heart failure: Prospective results from the Moli-sani study. Nutr Metab Cardiovasc Dis 28(3):298–307Google Scholar
- 187.Engelen L, Schalkwijk CG, Eussen SJ et al (2015) Low 25-hydroxyvitamin D2 and 25-hydroxyvitamin D3 levels are independently associated with macroalbuminuria, but not with retinopathy and macrovascular disease in type 1 diabetes: the EURODIAB prospective complications study. Cardiovasc Diabetol 14:67PubMedPubMedCentralCrossRefGoogle Scholar