Relationship between low vitamin D status and extra-skeletal diseases: a systematic review on effects of prophylaxis with vitamin D

Lorella Magnani

Hospital Pharmacy, Pharmaceutical Department, Tuscany North West Health Company, Area of Viareggio, Italy

Abstract

Key words: vitamin D extra-skeletal effects; vitamin D supplementation; vitamin D deficiency; vitamin D prophylaxis; vitamin D status;vitamin D low level; cholecalciferol; vitamin D panel

INTRODUCTION

Background

In the first decades after the discovery of vitamin D, studies on its effects were mainly focused on causal nexus and with the prevention of rickets and osteomalacia.In recent years,the increase in the prevalence of hypovitaminosis D, linked to the aging of population and lifestyle modification, has directed research on vitamin D, not only on its role in mineral and bone metabolism, but also on extra-skeletal effects.1-4A recent large body of observational evidence has suggested that lower vitamin D status has been linked to several health outcomes, including muscle-skeletal (rickets, bone fractures,osteomalacia, osteopenia, osteoporosis and muscle weakness) and non-skeletal complications (cardiovascular disease,pulmonary disease and cystic fibrosis, inflammatory bowel disease, cancer, diabetes, immune system diseases and risk factors), that has significantly increased the consumption of vitamin D supplements and laboratory tests.2,3,5To clarify the direction and strength of the association of low vitamin D levels on extra-skeletal outcomes, this systematic review of systematic reviews, meta-analysis, guideline and primary studies was conducted, as the most suitable study design for the synthesis in a single document about the available evidence of efficacy, considered the difficulty to extricate itself in the primary literature, whose discordant results and the very variable quality (size and characteristics of the sample, design,outcomes, statistics) are certainly not useful for evaluating the effectiveness of an intervention.This critical examination is not proposed to provide indications for health decisions, but to make us reflect on the inconsistency on which our knowledge is sometimes based: a starting point for reflection on the evidence available before making decisions on future research needs.

Objectives

The primary aim of this study is to summarize the evidence on a specific question, which is the association/causality of low vitamin D status with extra-skeletal disease, benefits and harms of vitamin D supplementation used in prophylaxis (for any formulation, dose and timing), coming from systematic reviews, meta-analysis, guidelines, primary studies, documents of scientific societies.The study also proposes to explore the heterogeneity of primary and secondary publications on the specific question, and highlight areas of uncertainty and stimulate research.

The research question was formulated according to the population, intervention, comparison, outcomes and study design method (Table 1).

DATA AND METHODS

Selection inclusion/exclusion criteria of the articles

The review includes randomized controlled trials (RCTs),observational studies, meta-analysis and systematic reviews,guidelines and documents issued by scientific societies,editorials and books concerning vitamin D low status and supplementation, focused on:

(1) Vitamin D levels: definition of deficiency/insufficiency and metabolism;

(2) Pharmacological supplementation with vitamin D (any product, dose and pharmacological form, timing and pharmacokinetics) for outcomes skeletal and extra-skeletal outcomes;

(3) Effects of vitamin D low status on extra-skeletal outcomes;

(4) Adults of both sexes, excluding pregnant women and children, individuals with bone disease (e.g., osteoporosis,osteopenia, rickets).

Furthermore, non-English language publications and literature unpublished have been excluded.

Sources of evidence

The Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) statement was used for reporting the present systematic review.The Articles indexed in PubMed, MEDLINE, Cochrane Library were searched using the following MeSH terms [title or abstract]: “Vitamin D insufficiency”; “Vitamin D deficiency”; “Vitamin D low level or status”; “extra-skeletal effects or outcomes of Vitamin D”; “skeletal and extra-skeletal effects or outcomes of Vitamin D”; “supplementation with Vitamin D”; “supplementation with cholecalciferol”; these terms was looked for in the abstract, title, or keywords.After the eligible full-text articles were reviewed and the relevant data reported in those articles were further searched, the following information was extracted from each: the first author, year of publication, type of study, number and characteristics of the population (cases and controls), unit and method of vitamin D measurement in serum, outcomes and results; for vitamin D supplementation:form, dosage and duration of treatment, assessment duration,methods and parameters for outcome measurement, vitamin D status at baseline and after vitamin D treatment, and treatment outcomes.To evaluate the state of the art, RCTs, observational studies, meta-analysis and systematic reviews, guidelines and documents issued by scientific societies, editorials and books concerning vitamin D and supplementation were examined,for a total of 48 references.The synthesis of the results is prevailingly narrative because of the heterogeneity within the same revision and for the different revisions; given that the included studies come from different settings, it is reasonable to believe that the conclusions can be generalized also to different clinical contexts and socio-cultural realities.

Research selection

The review of the literature on the extra-skeletal effects of vitamin D and relative supplementation was conducted, in agreement with PRISMA statement, by searching in PubMed,Cochrane Library and Medline; the main selection criteria used has been the following: studies (RCTs, observational,meta-analysis), and systematic reviews, focused on the low vitamin D status specifically on risk factors and extra-skeletal disease and on the relative effects of vitamin D supplementation (any type of vitamin D and any dose with any duration and route administration versus placebo or no intervention),in adult participant (excluding children and pregnant women),published from 2000 to 2018 in English.Publications that met the (including/excluding) criteria were examined considering the information gathered for each study primarily on type of goal, target population and setting, design and execution,duration and follow-up, management bias, findings and also noted conclusions drawn by the investigators and a narrative synthesis was made.The review included 18 systematic reviews/meta-analyses, 11 documents of scientific societies/guidelines/editorials/books, and 19 interventions and observational studies (of which 5 RCTs).

Methods

The selection process of the studies started with the qualifications, verifying that their content contains elements relevant to the question and the criteria for default inclusion or exclusion;subsequently same process has been applied to the full article.In reviewing the content of the selected articles, in order to extract elements specifically related to the review issue (effects extra-skeletal of the vitamin D and supplementation) books,guidelines and documents issued by the scientific societies of the sector were consulted, delimiting the topic to key concepts:synthesis and metabolism of vitamin D, definition of deficiency/deficiency.Later other key concepts: effects of low vitamin D concentrations on extra-skeletal outcomes and supplementation, working with a schema-concept map connected to the question for extracting the following information:

- Studies characteristics: design, objective, inclusion/exclusion criteria, outcomes;

- Characteristics of the participants: age, gender, comorbidity,and others relevant to the question of the revision;

- Intervention/exposure and setting;

- Measured outcomes and bias: definition and tools used to measure them;

- Results and conclusions.

The results chapter summarized factors to be considered in the evaluation of the vitamin D status and in the supple-mentation deduced from the examined literature, which are“suggestions” and not “directives.”

RESULTS

Clinical issues

Physiology and metabolism of vitamin D

Vitamin D is a hormone with both skeletal and extra-skeletal effects; exists in two isoforms, vitamin D3(cholecalciferol)whose needs derive mainly from skin synthesis (90-95%)and vitamin D2(ergocalciferol) mainly acquired through diet.The main physiological function of vitamin D is to maintain calcium and phosphorus extracellular concentrations within a normal range, both maintaining the efficiency of the small intestine in absorbing calcium and phosphorus in the diet,and stimulating the mobilization of calcium and phosphorus reserves from the bone.When the skin is exposed to sunlight,the ultraviolet B rays (between 290 and 315 nm) induce the conversion of the 7-dehydrocholesterol to provitamin D3,which in turn can be isomerized in vitamin D3or can absorb other ultraviolet B rays and isomerize into photoisomers inactive biologically, lumisterol, tachisterol, suprasterol I and II (which are the probable explanation for which the vitamin D intoxication does not occur due to chronic exposure to sunlight); the other form is vitamin D2(ergocalciferol)mainly acquired through diet.6Vitamin D, absorbed by the diet or synthesized in the skin for exposure to ultraviolet B rays, is stored in the liver and adipose tissue.The first stage of activation occurs in the liver where it is hydroxylated to 25-hydroxyvitamin D (by CYP2R1) and transported to the kidney linked to the vitamin D-binding protein (VDBP), also known as Gc protein.The second stage of activation occurs in the kidney (mainly in the proximal tubule) where it is hydroxylated (by CYP27B1) to 1.25-dihydroxy vitamin D and subsequently transported to the target tissues where it binds to the nuclear vitamin D receptor (VDR).6The pharmacokinetics of vitamin D is therefore considerably more complex than a pharmacological agent standard, due to the necessary hydroxylation in the active form, 1,25(OH)D and it is binding to the plasma binding protein VDBP, which influences the properties of metabolites.6One aspect that should not be underestimated is that the enzymatic system used for this biosynthetic process becomes less efficient with aging (e.g., intestinal absorption),which explains the high frequency of hypovitaminosis D in the elderly population, and that is strongly influenced by the environment, from nutrition and genetic determinants, which explains the difficulty in determining blood concentration and in finding common consensus.3,6-8

Actions of vitamin D on skeletal and extra-skeletal outcomes

Vitamin D is biologically inert and requires two essential hydroxylations to convert to its active form, the 1,25 dihydroxycholecalciferol [1,25(OH)2D3].When calcium intake in the diet is insufficient, the 1,25(OH)2D3, together with the parathyroid hormone, acts on the osteoblastic cells by stimulating the expression on their cell surface of the receptor activator of nuclear factor kB, which induces monocytes to differentiate into osteoclasts: the interaction of the receptor activator of nuclear factor kB with its receptor induces the process of bone remodelling.9,10However, although it is known that the 1.25(OH)2D3is involved in the endocrine system of bone metabolism (intestinal absorption and mobilization from calcium and phosphate bone), there are no evidence that 1,25(OH)2D3directly stimulates bone formation and mineralization.9-11The expression of the VDR in most nucleated cells (although frequently at a low level), suggested that the endocrine system of vitamin D may have a much broader set of activities than just calcium/ bone homeostasis and encouraged to conduct studies on extra-skeletal outcomes.2In vitroand pre-clinical studies have revealed that the cells of the immune system express the VDR, that almost all the cytokines involved in the immune system are under the control of 1,25(OH)D, that the endocrine system of vitamin D has action on monocytes and macrophages (innate immunity activation) and on helper T cells (inactivation acquired immunity).2On the basis of these results, RCTs were conducted to verify whether hypovitaminosis D is associated with a higher risk of infections, but the conclusions are very discordant.A recent meta-analysis12found that vitamin D supplementation reduces the risk of upper respiratory infections, but insufficient to define the dose and population at risk.The possible link between lower vitamin D status and risk of cardiovascular events is based on the control by 1,25(OH)D on some fundamental genes for cardiovascular function (e.g., renine, trombomodulin), but a large number of studies (case-control, observational, metaanalysis) found a low causality of low vitamin D status and increased risk of cardiovascular events.Moreover, it has been observed that high concentrations of 1,25(OH)D can stimulate the trans-differentiation of vascular smooth muscle cells similar to osteoblasts and induce ectopic calcifications.The vitro and preclinical data are suggestive for a wide spectrum of extra-skeletal vitamin D activity; however, most RCTs have produced null or contrasting results for most of these outcomes, so the Institute of Medicine (2010) and others scientific societies2,4,13,14have proven the suspicion about the effects of vitamin D supplementation on the immune system,cardiovascular and metabolic diseases (e.g., diabetes).The primary prevention study VITAL,15,16large and long lasting (≥5 years), with sufficient power to examine the effect of vitamin D (2000 IU/d) and omega-3 (1 gr/d) supplementation on the risk of invasive cancer of any type or major cardiovascular event (myocardial infarction, stroke and cardiovascular death)compared to placebo did not show a significant benefit of vitamin D supplementation on the primary endpoint: the supplementation did not result in a lower incidence of total cancer deaths or a lower incidence of breast, prostate or colorectal cancer than placebo and not even a significant difference in the secondary cardiovascular endpoints or death rate for any cause in the total cohort or in the subgroups.At the moment,the evidences that deserve further investigation concern the benefits on infections of upper respiratory tract and a reduction in the number of exacerbations in chronic obstructive pulmonary disease (COPD), using 2000 IU/d supplementation, even if do not derive from comparative studies.17Although vitamin D deficiency has a high prevalence in COPD and other lung diseases, due to multiple causes (limited physical activity and therefore less sun exposure, systemic inflammation, incor-rect nutrition, corticosteroid intake, genetic polymorphisms),the question is that it is a consequence rather than a direct cause.2,18,19Indeed, although a significant association between pulmonary function decline and low levels of 25(OH)D has been observed in several prospective cohort studies and in genetic studies, it has been found that polymorphisms in the VDR gene are risk for lung diseases, no causality has been demonstrated by intervention studies: no correlations between ventilator and gas parameters have been demonstrated with 25(OH)D concentrations if not related to the pathophysiology of the disease.19Moreover, the discussion of which strategy should be used for supplementation is still ongoing, although a systematic review with meta-analysis by Martineau et al.12showed that patients receiving daily doses benefit from vitamin D supplementation, while no effect was found when given at high bolus doses.

Vitamin D panel

The main reason for the controversy in the scientific community concerns the evaluation of the vitamin D status (vitamin D panel), since the correct determination has important consequences on the diagnosis of deficient states and relative correction, as well as on the dose supplementation.2,7,11,20To monitor vitamin D status, it is used commonly active metabolite, the 25-hydroxycholecalciferol [25(OH)D] produced in the liver by hydroxylation (by CYP2R1) of two isoforms, one endogenous (Vitamin D3) due to cutaneous synthesis by exposure ultraviolet B and an exogenous (vitamin D2) due to dietary intake.Although the biological variability of short-term of 25(OH)D (6 weeks about)is less than 7%, it is referred to as “difficult analyte” commonly,due to changes in concentrations, on average 40 nM during the year, with variations peaking up to 105 nM in relation to genetic factors (gender, ethnicity, polymorphisms) and environmental factors (exposure to sunlight, diet, direct or direct integration of vitamin D, skin pigmentation, physical exercise), sensitivity and specificity of laboratory method.To address this difficulty, in 2010, the National Institutes of Health organized a standardization program of vitamin D to define which criterion to perform a coefficient of variability ≤ 10%.7,11,21-26In recent years, there has been much discussion about the “normal range” of 25(OH)D and, at present, while important differences remaining, both the Institute of Medicine and the US Endocrine Society have come to the same conclusion that a blood level of 20 ng/mL (50 nM)represents the acceptable limit beyond which the physiological functions guaranteed are: absorption of calcium from the intestine and control of the levels of parathormone.22However,while the experts of the Institute of Medicine have identified 16-20 ng/mL (40-50 nM) concentration that guarantees the physiological functions of 25(OH)D23in the general population and alerted to the potential adverse effects at long-term of serum levels ＞ 50 ng/mL (＞ 125 nM), due to lack of data, the US Endocrine Society, based on studies of the pharmacodynamics of calcium and bone20,24-26and referring to elderly adults in a more specific way, produced a guideline where it introduced the “range of deficit” to 20-29 ng/mL (50-74 nM), shifting the “range of normal” to 30-100 ng/mL (75-250 nM).Noting the above differences in the cut-off of 25(OH)D20and that the conservative perspective of the limit of 50 ng/mL may fail to minimize seasonal fluctuations, in the decision making treatment process, it is important to consider that, applying the“range of insufficiency” indicated by the Endocrine Society guideline to the adult general population, most would be a candidate for supplementation6: in this regard, it is interesting the explanation of the authors of the study published in 2016 on NEJM “Vitamin D deficiency-Is there really a pandemic?”about the interpretation that the entire population must have a blood value ＞ 20 ng/mL, both the reason for the increase in expensive laboratory tests and over-treatment27; in any case,the main scientific institutions believe that the threshold value＜ 20 ng/mL (frequent in the elderly population) reflects unfavorable condition for skeletal health.Therefore, given the discrepancy between the “range of normal,” mainly due to a different interpretation of the results of clinical trials, from a diagnostic point of view, it could be misleading to take a single measurement of 25(OH)D, in any moment of the year,to extrapolate the “patient’s vitamin status,” without considering the annual rhythm of the metabolite in relation to age(specially), genetic factors (gender, ethnicity, polymorphisms of the VDBP gene and VDR), environmental (exposure to sunlight, diet, body mass index, intake of calcium, directindirect integration of vitamin D, skin pigmentation, physical exercise).20For these reasons, predictive models that take into account these variables are studying; another factor not to be neglected is also the analytical variability7: although the main producers of diagnostic systems have recently improved the automated dosages for the determination of 25(OH)D, the intra-inter-laboratory variability is still higher than the desirable criterion of coefficient of variability ≤ 10%, especially at low concentrations and therefore lead to misdiagnosis of deficiency/insufficiency.3

Therapeutic issues

Effects of vitamin D supplementation on skeletal and extra-skeletal outcomes

A recent meta-analysis29extensively debated internationally,concluded that vitamin D supplementation does not reduce the risk of fractures compared to placebo or no treatment in subjects over 50 years old not institutionalized, not affected by osteoporosis or other pathologies of the bone, without previous fractures or in treatment with drugs that act on the bone metabolism.The same conclusions were derived from the United States Preventive Services Task Force in the update to the 2012 recommendations30where it is reported that there is insufficient evidence to support that supplementation brings net benefits for the prevention of falls in subjects over 65 years living in community, who do not have osteoporosis or deficiency of vitamin D or previous fractures9not even of generalized screening for the determination of hypovitaminosis,since there is no unanimous definition of “range of normality”and an adequate standardization of the laboratory parameters.10With regard to the effects of supplementation on bone mineral density, the authors of the meta-analysis published in Lancet in 201421,27,31conclude that continuous use and widespread for the prevention of osteoporosis in adults living in the community,without specific risk factors for “deficiency,” is inappropriate,considered that the densitometric effects of vitamin D are moderately modest, proportionate to the degree of deficiency and biphasic.31-34A recent work35has confirmed or not the role of hypovitaminosis on the risk of fractures with Mendelian randomization; that is, the approach that uses genetic polymorphisms to explore the effects of modifiable exposure factors on outcomes of interest, in the presence of confounding factors.When measuring vitamin D levels, it is now clear that many other factors need to be considered (physical activity, diet, sun exposure, skin pigmentation, age, body mass index) in addition to the effects on serum vitamin D levels, and that may all independent on the risk of fractures and to determine errors of assessment8.Therefore, the rationale for the use of Mendelian randomization lies in the fact that it is not always clear which of the associated variables is the cause (inverse causality) or when both are joint effects of other variables (confounders).Even if statistical techniques allow to control/purify the effect of confounders, not all confounders can be considered or observed and the randomized clinical trials are not always feasible for practical reasons (e.g., very large samples and long observation times), for which the determination of the causality of the relationship remains affected by uncertainty.2,8,13,35-38In the design of a clinical study evaluating the effects of vitamin D, recruitment through the Mendelian model could be a solution to eliminate the confounders associated with both exposure factors of interest, selecting those subjects that are genetically predisposed to hypovitaminosis D, whose origin, by definition,is independent of environmental and dietary factors.Moreover,given that this type of randomization is not conditioned by behavior changes, it eliminates the problem of inverse causality,as well as the influence of other types of confounding, typical of observational studies.The results of this interesting study conducted on genetic data of 500 thousand people, support the evidence that hypovitaminosis does not have an independent causal effect on the risk of fractures and reveal that, also all the other causal factors studied (age, diet, exercise physical, body mass index) are associated with fractures, but without causal relationship, except for bone mineral density, the only risk factor with strong evidence of causality.35At last, a systematic review36about the effects of high-dose vitamin D supplementation (＞ 800 IU/d) on fractures, falls and bone mineral density,which used meta-analysis published from September 2017 to February 2018, concluded that integration does not prevent fractures or falls, nor does it have significant effects on bone mineral density, showing that there are no differences between lower or higher doses.Many observational studies have found an association between vitamin D deficiency and muscle strength and increased risk of falls, but intervention studies have shown a modest benefit of supplementation in subjects with severe baseline deficiency.According to the review39and a subsequent meta-analysis,31using the same data, the effects of vitamin D supplementation on falls are found within the“useless border less than 15%.”40Waiting for further studies,the provisional conclusion appears to be that supplementation in vitamin D deficient elderly subjects could modestly decrease the risk of falls (about 20%) while serum increases above 125 nol/L or boluses of 300.000 IU should be avoided, as this strategy could have opposite results.2,13,14As for supplementation in the COPD, since most of the studies conducted are undersized and heterogeneous in terms of population, disease staging (Global Initiative for Chronic Obstructive Lung Disease criteria), drug and dosing regimen used, route of administration, seasonality, and corticosteroid intake.It seems reasonable to resort to supplementation not so much for the reduction of exacerbations (where there is no robust evidence) as for the correction of low levels of 25(OH)D generally associated with COPD comorbidity, such as osteoporosis.

Pharmacology of vitamin D supplementation (monitoring)

The cholecalciferol (vitamin D3) is the most effective active ingredient for treating deficient states and increasing the levels of 25(OH)D, preferably with daily dosage, avoiding (if not in specific conditions, as alterations to hepatic and renal metabolism) the hydroxylated metabolites of vitamin D (calcipediolcalcitriol-alfacalcidiol-ergocalciferol-dihydrotachisterol) as there is no evidence of efficacy similar to those available for cholecalciferol, comparative studies of equivalent doses are missing and because of an increased risk of hypercalcemia.Furthermore, the results of a recent study confirm that vitamin D3 (at the dose of 600 IU/d) is doubly effective in reestablishing the biological marker, of the vitamin D [25(OH)D]status compared to vitamin D2.28Opinions are also divergent in estimating the dose of supplementation37: if the routine generalized use is not indicated and does not appear economically acceptable, the 25(OH)D serum dose may be considered a good indicator in the following cases9,10:

· Conditions caused by and/or associated with a risk of deficiency23;

· Conditions of possible deficiencies related to the use of drugs (e.g., anticonvulsants, glucocorticoids, antivirals,antifungals, immunosuppressants, cholestyramine);

· Search for deficient states (diagnosis of bone diseases,renal and hepatic failure, malabsorption syndromes, hyperparathyroidism);

· Assessment of maintenance dose after cumulative treatment dose for correction of the deficiency status (to be performed after 4 months, time required to reach the study state serum of the metabolite)27;

· In subjects greater than 70 years, in cases of doubt and comorbidity;

· In cases of treatment with cholecalciferol at the doses greater than 4000 IU/d.

The most physiological approach to supplementation with vitamin D is the daily one: according to the Institute of Medicine (Recommended Dietary Allowance), once the need has been ascertained, in relapse prophylaxis (in subjects where the deficiency has been corrected) and in subjects with risk,the recommended dose of cholecalciferol to maintain levels of 20 ng/mL (50 nM) is 600-800 IU/d, which may be increased(according to the US Endocrine Society) up to 2000 IU/d,depending on the clinical situation and prefixed objectives;the supplementation of 500-800 IU/d was also shown to be effective in counteracting the bone mass loss of glucocorticoidinduced.23,27It is good to keep in mind that, the blood cut-off of 20 ng/mL is not the limit below which to define a “deficit”(the Institute of Medicine has estimated an average requirement of 16-20 ng/mL), but the limit to ensure the physiological functions of the vitamin D in the general population and that excessive supplementation, due to saturation of the VDBP,leads to increased metabolites [25(OH)D and 1.25(OH)D]and the respective “free vitamin D”fractions, that activate the VDR in an unregulated way, with consequent risk of toxicity(hypercalciuria, hypercalcemia, soft tissue calcification and possible renal damage and vascular).41Given the above, for the estimation of the therapeutic dose and maintenance according to the levels of vitamin D, it is suggested, in the current state,to follow the recommendations contained in the Guidelines.6The Institute of Medicine and most of the guidelines, to avoid the toxic effects of vitamin D (hypercalcemia), they consider not to exceed 4000 IU/d and 100 nM in the serum.11,20,27

Pharmacological supplementation with cholecalciferol:warnings:- In the switches between analogues of vitamin D, it is useful to consider the half-life, to avoid hypervitaminosis- Avoid “annual boluses”42-44 and doses greater than 100,000 IU due to the risk of increased bone resorption (biphasic effect), consider that the 100,000 IU formulations expose to the risk of overdose32- Do not exceed the maximum cumulative dose of 600,000 IU/year- Limit the fortnightly/monthly dosing (doses greater than 50,000 IU) to situations in which treatment adherence cannot be guaranteed [up-regulation of CYP24A1 and reduction of levels of 1,25(OH)D]6,27- Limit the intramuscular administration to patients with malabsorption, because oral absorption is higher27,45,46- Do not exceed concentrations of 25(OH)D ＞260 nM due to the risk of hypercalcemia (＞ 4000 IU/d)23- Consider that the 25(OH)D response to a specific dose of supplementation is independent of age and calcium intake, the opposite for high body mass index6,27- Consider that calcium, phosphorus and PTH influence the half-life of 25(OH)D, which varies from 10-40 days6- Avoid supplementation for extra-skeletal outcomes (neoplasms,psoriasis, autoimmune diseases, diabetes, cardiovascular disease/mortality, multiple sclerosis, dementia, chronic obstructive pulmonary disease) because evidence of efficacy is scarce and dissimilar6,27,28,45

DISCUSSION

Summary of evidence

This review draws attention to the question that, although there is a large amount of observational cohort studies and anecdotal evidence of the benefits of vitamin D supplementation,more research is needed to apply these results in a widespread way: while there is little solid evidence from RCTs or other studies of intervention, there are results of several studies cancel each other out to opposite conclusions and pervasive methodological flaws that render the evidence unreliable.The motives of inconsistencies between the studies cited in this review could be due to the difference of vitamin D initial status, length of follow-up, patient compliance, difference of drug and dose for supplementation, sample size, population characteristics and heterogeneous metabolism among the included rate patients (e.g., diversity by gender, age, race,lifestyle).Therefore, at present, the definition of “vitamin D deficiency”and the indication to supplementation and relative dose, are still problematic, even if some consensus points have been identified and translated into guidelines of Italian Society of Osteoporosis, Mineral Metabolism and Skeletal Disease.47Given that the causal relationship between “vitamin D status” on skeleton and extra-skeleton should be made up with randomized clinical trials, difficult to plan for many reasons(already mentioned), a new design opportunity could derive from the genetic epidemiology that uses Mendelian randomization, which eliminates the confounding factors typical of observational epidemiology, which often generate solid but non-causal associations.8However, we must consider that,even this randomized method has some limitations, such as the confounding of other polymorphisms linked to the one in question or other sources of variation that generate inaccuracies and non-reproducible data outcomes could come from Mendelian randomization, which eliminates the confounding factors typical of observational epidemiology, which often generate robust but non-causal associations.Anyhow, the demonstration of causality should derive from RCTs, since,even this randomized method has some limitations, such as the confounding of other polymorphisms linked to the one in question or other sources of variation that generate inaccuracies and non-reproducible data; indeed, although the single nucleotide polymorphisms involved in vitamin D metabolism(cytochrome P450 and VDBP enzymes) and/or VDR are the ones that received the most attention, genome-wide association studies have shown that even genes not directly involved in activation of the VDR are significant determinants of the vitamin D status (such as Kruppel like factor 7-protein coding gene and phosphatidic acid phosphohydrolase 1 gene involved in the hydroxylation-metabolism of vitamin D and its metabolites).The polymorphisms in the Gc gene that codes for VDBP, of which there are racial and geographic differences in the predominant frequencies, are those that have a greater association with the risk of vitamin D deficiency and in the modulation of the supplementation response, but at the moment, it is not clear if the Gc genotype directly influences the circulating concentrations of vitamin D metabolites, its metabolism or whether the effect is mediated through the plasma concentration of VDBP; even this interesting alternative study design still requires in-depth studies on which it is worth investing.6-8Ultimately, considering that:

· A causal relationship of vitamin D has not been demonstrated on fractures, bone mineral density and extra-skeletal effects,both due to the difficulty of study planning (design, duration,enrolment, confounding factors, costs) and due to the complexity of the chemistry and kinetics of the vitamin itself;

· An agreement has not yet been reached on the “range of normal” of the metabolite 25(OH)D;

· The most recent evidence and opinions of experts suggest that vitamin D supplementation has no clinically significant effect on reducing fracture and or fall risk, on bone mineral density and extra-skeletal effects, the increasing use of supplementation, and with high doses, it does not seem justified, if not in clinically documented deficiency states or specially designed clinical trials, at the moment.

Limitations

This systematic review has some limitations: Heterogeneity due to clinical differences (participants, interventions,outcomes) and methodological (design and quality) and analysis taken into to the discussion and conclusions.Variety in the vitamin D supplementation regimens (contribute to unsettle treatment results); restriction to English language articles, inclusion of studies which took place at different points in time relative to published guidelines.Moreover,no web search was performed to retrieve any unpublished literature and the Assessing the Methodological Quality of Systematic Reviews score was not used to evaluate the quality of the reviews included.At last, it is not excluded that the revision, even if conducted with the PRISMA statement,may contain the selection bias (and extraction of results)introduced during the selection process, being conducted by a single investigator.

Conclusions

Implications for clinical practice

Vitamin D laboratory testing, prescriptions and costs associated with these practices have increased, in some cases dramatically, over the past 10-15 years.Ambiguity and inconsistencies in vitamin D management guidelines may explain the excessive number of repeat vitamin D tests and over or not recommended supplementation.

Implications for clinical research

Noted that current evidence presents a high risk of partiality,due to design, statistical power, inconsistent results and numerous other problems, then, to contain the enthusiasm for a“Vitamin D panacea”, in the near future, the research must be“re-oriented” to check the causality (and not only correlation)of low vitamin D status on extra-skeletal (and on skeletal)outcomes and therefore of the doses and timing for relative supplementation.

In particular, future studies should be planned with new designs to cope with the countless bias to which this area of research is subject, for example, using the genetics to design needful RCTs to study the causality of vitamin D status on extra skeletal outcomes, as well as the efficacy of pharmacological supplementation prophylaxis.

Better management of vitamin D may serve to achieve higher value care.

Additional file

Additional file 1: List of abbreviations.

Author contributions

LM is the sole contributor/author who drafted entire manuscript and completed the final edit/proofing review of the manuscript.

Conflicts of interest

None declared.

Financial support

None.

Reporting statement

This study followed the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidance.

Copyright license agreement

The Copyright License Agreement has been signed by the author before publication.

Data sharing statement

Datasets analyzed during the current study are available from the corresponding author on reasonable request.

Plagiarism check

Checked twice by iThenticate.

Peer review

Externally peer reviewed.

Open access statement

This is an open access journal, and articles are distributed under the terms of the Creative Commons Attribution-NonCommercial-ShareAlike 4.0 License, which allows others to remix, tweak, and build upon the work non-commercially, as long as appropriate credit is given and the new creations are licensed under the identical terms.