Introduction

Vitamin D plays a vital role in maintaining bone health. In addition, effects on inflammation, infection, allergy, cardiovascular function, and cancer have been reported [1,2,3]. Hypovitaminosis D is common in kidney disorders, and a large number of studies have reported on its osseous and extra-osseous effects in these conditions [4, 5]. This review focusses on examining the clinical relevance of low vitamin D status, as well as outcomes of native vitamin D (NVD) supplementation in children with kidney disease.

Online medical databases, Medline, Pubmed, and Google scholar were searched for articles related to vitamin D status, associations of hypovitaminosis D, and effects of vitamin D therapy in kidney diseases. The available data is summarized and inferences drawn from this data are presented as conclusions after each section.

Vitamin D physiology

Vitamin D is sourced from sun exposure and from diet. The term “native vitamin D”, refers to ergocalciferol (D2) which is chiefly available from plant sources; and cholecalciferol (D3), available from animal dietary sources, and produced from 7-dehydrocholesterol in the skin.

A specific vitamin D binding globulin (DBP) and to a smaller extent, albumin, transport the vitamin D2 and D3 to the liver. These compounds are hydroxylated in the liver to 25 hydroxy D2 and 25 hydroxy D3 (25[OH]D or calcidiol) and subsequently in the kidneys by 1α hydroxylase enzyme, to the active forms: 1,25 dihydroxy D2 and 1,25 dihydroxy D3, (1,25[OH]2D or calcitriol) which activate vitamin D receptors (VDR). The enzyme 1α hydroxylase is also found in non-renal tissues, therefore conversion of calcidiol to calcitriol can occur at a smaller magnitude, extra-renally. It is hypothesized that the calcitriol produced in the kidneys has exocrine functions, whereas extra-renal production is responsible for autocrine and paracrine effects, and not released into the circulation [6, 7].

Circulating serum total 25[OH]D (the sum of 25[OH]D2 and 25[OH]D3) is considered to be the best biological marker of an individual’s vitamin D stores. This is the most abundant of all vitamin D metabolites in the serum, levels correlate with the amount of NVD substrate available, and are stable for 3–4 weeks in the absence of proteinuria [8, 9]. In contrast 1,25[OH]2D has a short half-life, is rapidly influenced by calcium and parathyroid hormone (PTH) levels, has thousand-fold lower circulating levels and there are greater challenges of assaying. The latter has limited use in certain conditions such as chronic kidney disease (CKD), hereditary phosphate losing disorders, vitamin D–resistant rickets, etc. [8, 10].

The largest component of measured serum 25[OH]D (85 to 90%) is bound to DBP, while 10 to 15% is more loosely bound to albumin, and a small fraction: 0.03 to 0.04%, circulates in the free state [8, 11].

The DBP gene is the most polymorphic gene known, and DBP levels and vitamin D binding capacity may be influenced by genotype and be different between races. Evidence suggests that like thyroid hormones, sex steroids and cortisol, vitamin D follows the “free hormone hypothesis”. This proposes that “only the non-bound or free fraction of hormones that otherwise circulate in blood bound to their carrier proteins is able to enter cells and exert their biologic effects” [8, 10, 11]. This hypothesis is supported by several reports: a patient with homozygous DBP mutation, had nearly undetectable levels of total 25[OH]D but free 25[OH]D, calcium, phosphate and PTH levels were normal. Similarly, DBP knockout mice did not reveal any evidence of vitamin D deficiency despite very low levels of total 25[OH]D [11].

However, it is of note that in the proximal renal tubular cell, DBP–25[OH]D complex can be directly reabsorbed from the glomerular filtrate through megalin and cubilin pathways [8, 11].

NVD assays

Measurements of total 25[OH]D use various methods, including immunoassays (most common), protein-binding assays, high-performance liquid chromatography–ultraviolet (HPLC–UV), or liquid chromatography–tandem mass spectrometry (LC–MS/MS) [8]. Until recently, assay variation contributed to the confusion surrounding efforts to define hypovitaminosis D. The vitamin D standardization program (VDSP) in 2010 developed a reference measurement system for harmonized reporting through standardization of the above assays. The gold standard reference measurement procedure utilizes separate serum measurements for D2 and D3 using isotope dilution LC–MS/MS, as it has least interference. Immunoassays, on the other hand, suffer from matrix and other metabolite interferences [12]. VDSP implementation has seen drastic improvements in performance of assays of several manufacturers; with acceptable specifications defined by limits for total coefficient of variation (CV%) ≤ 10% and mean bias ≤ 5% [13].

Free 25[OH]D levels can be estimated with a formula that entails measurements of total 25[OH]D, DBP and albumin levels, but this suffers from inaccuracies mainly due to complexities and uncertainties of DBP measurement. An enzyme-linked immunosorbent assay (ELISA) kit commercialized by Diasource is available to directly measure free 25[OH]D. While direct measurement is definitely the better approach in terms of performance [14, 15], the lack of a reference method, interferences from lipids, bilirubin, hemoglobin, and the underestimation of D2 concentrations remain some limitations [16]. Nonetheless, studies have shown directly measured free 25[OH]D assay to have considerable significance in populations with altered DBP and albumin levels [17, 18]; however, to date, it is only available as a research tool.

Several other biomarkers and vitamin D metabolites are being tested for clinical validity; all except serum 25[OH]D warrant standardization [19].

Section A: Vitamin D in CKD

CKD is associated with bone, cardiovascular, and inflammatory abnormalities, which influence long-term survival [4]. The deficiency of 1α hydroxylase enzyme causes reduction in conversion of 25[OH]D to the activated form: 1,25[OH]2D contributes to the development of CKD mineral bone disease (MBD) [20, 21]. In addition, total serum 25[OH]D levels, are commonly reported to be low [22,23,24,25,26].

Causes are postulated to be poor appetite, malnutrition, reduced sun exposure and food restrictions, which are common in CKD. Deficiency, with 25[OH]D levels < 20 ng/ml, was reported in 28% of 506 children with estimated glomerular filtration rate (GFR) 30 to 90 ml/min/1.73 m2 at enrolment in the CKiD study [27], while levels < 16 ng/ml were found in 2/3 of 500 European children with CKD stages 3–5 [28] and in 27% of 29 children with kidney transplant [29]. 25[OH]D levels are reported to be lower as kidney function declines and are more marked in children with glomerular and proteinuric diseases [23, 28, 30, 31].

NVD and bone parameters in CKD

In early CKD, low 25[OH]D levels have been associated with secondary hyperparathyroidism (HPTH) even in the presence of normal 1,25[OH]2D [24, 32]. Other studies indicate that 25[OH]D levels correlate positively with estimated GFR, 1,25[OH]2D, and NVD supplementation, while negative correlations are reported with serum PTH, CKD stage, and FGF23 levels [23, 28, 30, 33, 34]. Vitamin D deficient children have been reported to have reduced height [26], as well as lower bone mineral density (BMD) scores in CKD stages 2–5D [35].

Effect of NVD supplementation

Clinical outcomes of NVD therapy, such as effect on growth, fracture risk, or bone biopsy, have not been studied in children with CKD. Instead, surrogate biochemical outcomes are used, such as serum 25[OH]D and PTH levels, where changes after intervention may be observed in the short term. A prospective cohort study including 537 CKD children, demonstrated that fracture risk was independently associated with higher PTH levels [36]. Thus, one of the aims of management of CKD MBD is to control secondary HPTH, although optimum PTH levels at different CKD stages are debatable [37, 38].

Interventional studies in adult CKD have reported statistically significant increase in 25[OH]D levels and reduction of PTH levels after NVD supplementation; however, the clinical impact with regard to improvement in secondary HPTH appears to be low [39, 40]. The meta-analysis by Bover et al. included 974 non-dialysis CKD patients (9 randomised controlled trials (RCTs) and 5 observational studies). In the intervention arm, the pooled increase in 25[OH]D level was 20.6 ng/mL while decrease in PTH levels was only 10.5 pg/mL [40].

The effect of NVD supplementation on HPTH is reported to be related to NVD dose and to stage of CKD. Ennis et al. in a cross-sectional analysis of 14,289 patients, showed that in CKD 3–5, increasing levels of 25[OH]D up to 42–48 ng/ml with supplementation resulted in progressively lower PTH levels [41]. The beneficial effect of NVD was seen more in early CKD, and less in stages 4–5/5D where activated vitamin D compounds were needed to control PTH levels even after 25[OH]D levels were optimised [42,43,44,45,46,47].

Pediatric studies are all relatively small, but several also report significant rise in 25[OH]D levels and fall in PTH levels after NVD supplementation [32, 48, 49]. Shroff et al., in the only pediatric RCT of 40 children [50], reported that fewer patients treated with NVD developed HPTH, with significantly longer time to development of HPTH. Target 25[OH]D levels were achieved less in higher stages of CKD.

Significant adverse effects of NVD therapy have not been reported with therapeutic doses in the pediatric CKD literature. This is in contrast to treatment with activated vitamin D compounds (alphacalcidiol, calcitriol, paricalcitol, doxercalciferol) where complications such as hypercalcemia are more common [51, 52].

NVD and non-osseus effects in CKD

Several adult studies have associated vitamin D deficiency with poor cardiovascular outcomes and with increased markers of inflammation in CKD [53]. Pediatric reports are relatively few. Small studies have shown that in children with CKD, levels of 25[OH]D correlate negatively with surrogate indicators of cardiovascular morbidity like aortic pulse wave velocity (PWV) and aortic augmentation index (AI) [54, 55]. Left ventricular mass index correlates negatively with 25[OH]D levels and positively with PTH levels, blood pressure (BP) and AI [22, 56]. Altemose et al. [57], reporting on 580 children from the CKiD cohort, documented 1.9 times higher odds for being anemic in hypovitaminosis D, even after adjusting for several covariates including GFR and use of erythropoetin or hematinics. Levels of 25[OH]D correlated positively with anti-inflammatory and negatively with pro-inflammatory biomarkers in children on chronic hemodialysis [58].

Shroff et al. [59] reviewed 167 children from the ESCAPE cohort and showed that patients with 25[OH]D < 20 ng/ml had higher diastolic BP. The 5-year actuarial kidney survival was 50% with 25[OH]D levels < 20 ng/ml and 75% with levels above 20 ng/ml at baseline. Kidney survival increased 8.2% per 4 ng/ml increase in 25[OH]D levels (p = 0.03), independently of other confounders. Adult studies also indicate that higher 25[OH]D levels are associated with improved survival in CKD. In a meta-analysis of 10 studies and 6853 patients [60], relative risk of mortality for every 10 ng/mL rise in 25[OH]D was 0.86 (95% CI 0.82–0.91).

Effect of NVD supplementation

In vitro and animal studies have shown that NVD supplementation improves cardiovascular endpoints and reduces inflammatory biomarkers in CKD [53, 61]. In children, one small RCT [62] reported a significant decrease in erythropoietin dose in the group treated with NVD compared to control. There are no other studies in children that examine whether intervention with NVD or improvement in 25[OH]D levels improves any non-osseus parameters of CKD. A systemic review found NVD supplementation to be ineffective in lowering BP in otherwise healthy children [63].

In adult CKD, meta-analyses of available RCTs to date, show no consistent improvement in cardiovascular outcomes, anemia, immune function, health-related quality of life scores, hospitalisations or all-cause mortality, with NVD therapy [64,65,66].

Conclusions (A): NVD in CKD

  1. a)

    Utilization of NVD to maintain vitamin D sufficiency status is useful in early CKD, to prevent or treat secondary HPTH, and it is initially preferred over activated vitamin D compounds. Higher doses of NVD may be required in advanced CKD stages to achieve 25[OH]D target levels.

  2. b)

    In advanced CKD, NVD is often insufficient to control HPTH. In such cases, once target 25[OH]D levels are reached, activated forms of vitamin D (alphacalcidiol/calcitriol) may be required for control of HPTH and CKD MBD.

  3. c)

    NVD supplementation may be considered to maintain adequate 25[OH]D levels, as the latter has been associated with improved non-osseus outcomes as well as overall survival. However, direct interventional studies to support such use are not yet available.

Section B: Vitamin D in nephrotic syndrome and other proteinuric kidney diseases

As discussed previously, the majority of serum 25[OH]D (> 99%) circulates bound to proteins. DBP has a molecular weight similar to albumin, and both are lost in urine in proteinuric diseases. In gross proteinuria, urinary losses of DBP bound to 25[OH]D may overwhelm proximal tubular reabsorption capacity via megalin/cubilin pathways [9,10,11].

Free 25[OH]D is defined as circulating 25[OH]D which is bound to neither DBP nor albumin. Several studies indicate that in health, free and total 25[OH]D levels correlate closely, suggesting that there is no need for assessing free 25[OH]D levels separately. However, this correlation was lost in several disease states including nephrotic proteinuria [9, 10, 18]. There are no studies to date that indicate the level of proteinuria at which this dissociation may occur.

NVD in nephrotic syndrome

In pediatric nephrotic syndrome (NS), reports have documented urinary DBP loss, with low levels of serum DBP and total 25[OH]D during and after NS relapse [67,68,69]. However, unlike in CKD, this had no association with abnormalities of corrected calcium, phosphate, alkaline phosphatase (ALP) or PTH [69,70,71]. The total 25[OH]D levels returned to levels similar to healthy controls after ≥ 3 months of attaining remission in steroid sensitive nephrotic syndrome (SSNS) [70, 72].

In contrast, free 25[OH]D levels, were found to be similar to healthy controls both in NS remission and relapse [18], and this may explain why corrected calcium, phosphate, ALP and PTH levels do not change significantly in SSNS, despite markedly low total 25[OH]D. Only total and not free 25[OH]D correlated significantly and negatively with urinary protein:creatinine ratios (Spearman’s rank correlation coefficient, rs =  − 0.42, p < 0.001 and 0.04, p = 0.6), while total 25[OH]D had a stronger correlation with serum albumin levels than free 25[OH]D (rs = 0.64, p < 0.001 and 0.21, p = 0.13) [18].

Effect of NVD supplementation

Children with NS are usually treated with steroids, and glucocorticoid-induced osteoporosis (GIO) is a considerable concern, especially if the patient is frequently relapsing or steroid dependent. The question is whether treatment with NVD will improve bone health in such patients.

In patients with SSNS on steroid therapy, two small RCTs over 2–3 months [73, 74], and a longitudinal study over 1.5 years [75] reported better BMD or bone mineral content (BMC) in groups treated with NVD and calcium; however, 25[OH]D levels were not measured in these studies. A subsequent RCT [71] showed no change in BMD or BMC over 6 months, despite marked improvements in serum total 25[OH]D status in the supplemented group.

An RCT by Rooney et al. [76] failed to show benefit of calcium and NVD supplementation on lumbar BMD in GIO of rheumatological diseases.

In NS, hypercalciuria was detected in 50% of patients who received NVD supplementation of 60,000 IU weekly for 4 weeks. All patients in the intervention arm were in relapse and had low total serum 25[OH]D (median 7.0, interquartile range 3–16.7) ng/ml at recruitment. These levels rose to a median of 54.45 (interquartile range 46.45–87.75) ng/ml after supplementation, when the patients were in remission [71]. The question arises of whether total 25[OH]D is a valid target for supplementation, given the evidence of DBP–25[OH]D complex loss in urine and low serum DBP levels in NS [67,68,69]. Free 25[OH]D levels were not checked in this study.

Studies in DBP knockout mice show that despite the absent DBP and very low levels of serum total 25[OH]D, these mice do not develop evidence of vitamin D deficiency unless they are fed an NVD-deficient diet. Extrapolating from this, it appears that DBP binding serves as a reservoir for vitamin D compounds [11]. Thus, in gross proteinuric conditions, where DBP levels are expected to be low, although calcium and NVD supplementation may not specifically treat GIO, achieving daily dietary reference intakes (DRI) for these compounds should be considered, to maintain bone health, as suggested in the forthcoming IPNA clinical practice recommendations, as well as other recent guidelines [77, 78].

NVD in other proteinuric diseases

Several studies of children and adult CKD have indicated lower total 25[OH]D levels in patients with proteinuria, with levels correlating positively with serum albumin and negatively with degree of protein loss [25, 27, 28, 30, 79]. A large population-based adult study revealed increasing prevalence of albuminuria with decreasing 25[OH]D levels [80].

These studies are cross-sectional in nature, therefore causality of the association cannot be proven, and the low total 25[OH]D may simply be an effect of increased urinary losses of protein-bound 25[OH]D. Serum or urinary DBP, free or bioavailable serum 25[OH]D levels have not been studied in these conditions. Pending further evidence, it may be logical to consider the use of NVD therapy above DRI, in such situations, if there is associated evidence of hypovitaminosis D such as low corrected calcium, phosphate and high ALP and PTH.

Experimental data suggests that NVD and VDR activators reduce proteinuria, possibly by suppressing the renin–angiotensin–aldosterone system [81]. In glomerular diseases like IgA and diabetic nephropathy, adult studies have indicated improvement in levels of proteinuria with VDR activator therapy [82,83,84]. However, no such evidence is available in children or with NVD supplementation. There are also no studies examining other non-osseus effects of NVD therapy in proteinuric disorders. Although it is attractive to consider that infection-related NS relapses may be reduced with NVD therapy, Banerjee et al. [71] failed to find any effect of NVD supplementation on relapse frequency; however, their study was not adequately powered to examine this effect.

Conclusions (B): NVD in NS and other proteinuric diseases

  1. a)

    In proteinuric conditions, serum total 25[OH]D levels may not reflect the biologically active fraction, and preliminary evidence suggests that free 25[OH]D may be a better guide. However, easy-to-use validated assays and further supporting evidence are required before this can become a clinically useful tool.

  2. b)

    In NS, although calcium and NVD supplementation does not specifically treat GIO, maintaining DRI for calcium and NVD is suggested to optimise bone health. Total serum 25[OH]D may be targeted in SSNS, ideally when in remission for over 3 months.

  3. c)

    In children, as yet there is no evidence that NVD supplementation reduces proteinuria or has any other positive influence on non-osseous outcomes in proteinuric diseases.

Section C: NVD therapy in kidney disease

Therapeutic targets

Ideally, the levels of 25[OH]D targeted should be sufficient to maintain both osseus and non-osseus functions of vitamin D. However, data for such optimum levels is lacking, particularly for non-osseus outcomes, and guidelines are available to address biochemical bone parameters only. In otherwise healthy children, the level at which bone accrual may be optimum, with effective absorption of calcium from the gut and maintenance of normal ALP and PTH levels, is suggested to be > 20 ng/ml in a recent global consensus [77]. This target is generally accepted for children with kidney disease and normal kidney function. It is recognised that in CKD, higher levels (> 30 ng/ml) are required to prevent secondary HPTH [37, 41, 85] (Table1).

Table 1 Vitamin D status: Consensus guidelines
Full size table

NVD supplements

NVD supplements are available in the form of ergocalciferol (D2) and cholecalciferol (D3). Studies in adults suggest that D3 is superior to D2 in raising both total and free 25[OH]D levels and in lowering serum PTH [86, 87]. An RCT in adults on HD, comparing high-dose monthly vitamin D2 versus D3, also reported that the latter resulted in higher 25[OH]D levels [88]. We could not find any studies comparing D2 to D3 supplementation in children with kidney disease. Pediatric studies to date have not demonstrated any difference between D2 and D3 supplements, whether given daily, weekly, or as single bolus doses [89,90,91]. Pediatric guidelines suggest that either D2 or D3 in daily dose regimens can be used; however, since D3 has a longer half-life, it is preferred if using intermittent dosing schedules [37, 77]. Calcefediol, an extended-release derivative of 25[OH]D3 has recently been approved; however, there are no pediatric data available as yet.

NVD doses and monitoring

The specific dose required in an individual patient to raise 25[OH]D levels from a certain baseline to target levels is not known. Evidence-based pediatric guidelines that are not specific for kidney disease, recommend minimum daily intakes of 400 IU in infancy and 400–600 IU in older children. For treatment of rickets or low 25[OH]D status, the recommended daily doses are up to 2000 IU for infants or 3000 IU for older children; for 12 weeks. For reasons of improved compliance, intermittent doses of 50,000 to 60,000 IU weekly may be given to children over 1 year of age, with total cumulative dose remaining the same [77, 78]. These guidelines are usually followed also for non-CKD renal diseases. Since complications occur more frequently with mega-dose Stoss regimens, such treatment is best avoided [37].

Interventional studies in children with kidney disease have used D2 or D3, in widely variable doses and frequencies, varying from megadoses of 300,000 to 600,000 IU [49, 92] to 60,000 IU weekly for 4 weeks [71] to 2000 IU/day [93]. However, although median 25[OH]D levels increased significantly, the above-mentioned target levels were not achieved in all children with these doses.

It is recognized that higher doses may be required in CKD. Shroff et al. administered D2 supplements to children with CKD, in a modified DOQI protocol [50, 85], which varied according to age and baseline levels (Table 2), and reported that after intensive therapy, target levels of > 30 ng/ml were achieved in 80%, which fell to 60% on maintenance therapy.

Table 2 Guidelines for NVD replacement in pediatric CKD
Full size table

Achievement of target levels was less in higher stages of CKD, where even higher doses may be required. In patients with low vitamin D status, Kari et al. administered 2000 IU NVD daily for 3 months, and only 9% of patients achieved levels > 30 ng/ml [93]; whereas doses of 3000 to 4000 IU daily were able to achieve target levels in > 70% patients in two recent RCTs [31, 94].

The doses of NVD used in studies or recommended in guidelines are not generally scaled to body weight or surface area. However, nephrocalcinosis occurred more frequently in smaller children after excessive intake in a recent case series [95]. Pharmakokinetic simulation modelling has suggested that weight-based dosing may improve attainment of target levels and avoid toxicity due to overdose [96].

Screening for hypovitaminosis D is suggested in at-risk patients, every 6–12 months. Continued monitoring of levels at regular intervals while on treatment (Table 2) is required to guide further therapy, along with monitoring for toxicity (hypercalcemia, hyperphosphatemia or hypercalciuria) [37]. Adverse effects are more likely during high dose intensive replacement/loading regimens, and monitoring is more important during this time.

Toxicity of NVD

Pediatric studies using therapeutic doses of NVD reported increase in FGF23 in CKD [31, 97] while calcium, phosphate, and ALP levels were largely unchanged.

However, hypercalciuria was reported in > 50% of patients receiving NVD in NS, when supplementation was guided by total and not free 25[OH]D levels [71].

A meta-analysis of adverse effects of long-term (≥ 24 weeks) NVD supplementation in adult studies revealed increased risks of hypercalcemia and hypercalciuria [98].

Hypervitaminosis usually occurs due to ingestion of excess vitamin D supplements [94, 98]. Vitamin D toxicity is manifested as hypercalcemia, hypercalciuria and suppressed PTH levels. This may cause further complications like pancreatitis, nephrocalcinosis, nephro-urolithiasis, and kidney function impairment. There may be no early symptoms, and biochemical monitoring only can pick up early toxicity. Symptoms, when they occur, maybe be non-specific, like anorexia, vomiting, constipation, polyuria, polydipsia, and dehydration. Rare life-threatening conditions like arrhythmias or encephalopathy have also been reported [99, 100].

In the general population, 25[OH]HD levels > 50 ng/ml are associated with increased mortality, while symptomatic toxicity is associated with 25[OH]D levels above 100 ng/ml [37]. Guidelines suggest that vitamin D supplements are stopped before 25[OH]D levels of 50 ng/ml to prevent toxicity [37, 77].

The availability and formulations of vitamin D preparations are very variable even within countries and regions, and can often be bought as over-the-counter medications. The supplements are available in the form of very concentrated drops, syrups, powders, and capsules, as single products or in combination with calcium and other micronutrients. Popular media have generated a perceived need for NVD supplementation for a myriad of conditions and for general well-being. However, globally, there are now several reports of concentration errors during manufacturing, errors in labelling, and of inappropriate or wrongly administered doses [95, 101,102,103,104,105]. Public education, review and standardisation of formulations are required to safeguard against such mishaps. Caregivers administering these medications should be adequately counselled and treatment monitored to avoid toxicity. Availability without prescription should not be allowed for the more concentrated or high dose formulations.

Conclusions (C): NVD therapy in kidney disease

  1. 1.

    Patients who have high risk of hypovitaminosis D, including patients with hypocalcemia, HPTH, CKD, proteinuria, osteoporosis or clinical rickets, should be screened for vitamin D status.

  2. 2.

    In such patients, vitamin D sufficiency targets, i.e. total 25[OH]D levels of > 20 ng/ml in non-CKD and > 30 ng/ml in CKD, but < 50 ng/ml (in absence of gross proteinuria) should be achieved with NVD therapy (Table 1). Treatment is started as per available guidelines, but subsequently titrated according to levels of 25[OH]D achieved, as well as serum and urinary calcium and PTH status.

  3. 3.

    Supplements containing D2 or D3 may be used; the latter is preferred when using weekly intermittent therapy. Megadose Stoss therapy is to be avoided.

  4. 4.

    Regulations for standardization and labelling of formulations, parental guidance and monitoring are required, and concentrated or high dose preparations should only be available by prescription.

Summary

Although there is much discussion as well as published guidelines regarding NVD therapy in CKD, the issues surrounding proteinuric renal diseases are more nebulous.

In children with CKD, observational studies associate low 25[OH]D status with poorer osseous and non-osseous outcomes. However, evidence of benefit of NVD therapy is restricted only to biochemical improvements in 25[OH]D and PTH levels. Recent evidence-based guidelines aim to optimize NVD therapy to attain vitamin D sufficiency status.

In NS and other hypoalbuminemic/proteinuric conditions, serum total 25[OH]D levels may not reflect the biologically active fraction. Very few studies have correlated outcomes to free 25[OH]D levels. Until commercial assays and further studies examining free 25[OH]D are available, total 25[OH]D levels may be used to guide therapy when the proteinuria has remitted for ≥ 3 months; or NVD supplementation (above DRI) may be indicated in the presence of effects of true hypovitaminosis D, such as low corrected calcium, phosphate and high ALP and PTH levels.

Further research is required to assess the benefits of NVD therapy on clinical outcomes. The limitations of available assays in certain clinical situations needs to be recognized. NVD formulations should be regularised, and therapy monitored and supervised adequately to avoid toxicities which can cause acute systemic illness as well as long-lasting kidney impairment.