Vitamin D Supplementation in Children and Young Adults with Persistent Proteinuria Secondary to Glomerular Disease

Abstract

Background:

Vitamin D deficiency is common in glomerular disease. Supplementation may be ineffective due to ongoing urinary losses of vitamin D binding protein. We sought to determine if daily cholecalciferol supplementation would increase vitamin D concentrations in children with glomerular disease and persistent proteinuria, without adverse effects.

Methods:

Eighteen participants at least 5 years of age with primary glomerular disease and urine protein:creatinine ratio ≥0.5 were enrolled from four pediatric nephrology practices to receive cholecalciferol supplementation: 4000 IU or 2000 IU per day for serum 25 hydroxyvitamin vitamin D (25OHD) concentrations <20 ng/mL and 20 ng/mL to <30 ng/mL, respectively. Measures of vitamin D and mineral metabolism were obtained at baseline and weeks 6 and 12. Multivariable generalized estimating equation (GEE) regression estimated mean percent changes in serum 25OHD concentration.

Results:

Median baseline 25OHD was 12.8 ng/mL (IQR 9.3, 18.9) and increased to 27.8 ng/mL (20.5, 36.0) at week 6 (p<0.001) without further significant increase at week 12. 31% of participants had a level ≥30 ng/mL at week 12. Supplementation was stopped in two participants at week 6 for mildly elevated calcium and phosphorus, respectively, with subsequent declines in 25OHD of >20 ng/mL. In the adjusted GEE model, 25OHD was 102% (95% CI: 64, 141) and 96% (95% CI: 51, 140) higher versus baseline at weeks 6 and 12, respectively (p<0.001).

Conclusion(s):

Cholecalciferol supplementation in vitamin D deficient children with glomerular disease and persistent proteinuria safely increases 25OHD concentration. Ideal dosing to fully replete 25OHD concentrations in this population remains unknown.

Clinical Trial:

NCT01835639

Graphical Abstract

Introduction

Nephrotic syndrome is the most common acquired cause of chronic kidney disease (CKD) in childhood. It imposes a heavy burden of cardiovascular[1, 2], metabolic[3, 4], hematologic[5] and infectious complications.[6] Vitamin D deficiency impacts bone and mineral metabolism, cardiovascular disease, insulin resistance, autoimmunity and inflammation [7–11] and is highly prevalent in nephrotic syndrome, identified in up to 100% of patients. [11–13] Furthermore, animal studies have shown that activated vitamin D treatment reduces albuminuria and prevents podocyte loss and glomerulosclerosis. [14] Vitamin D therapy has also been shown to reduce proteinuria in adults with CKD.[15, 16] Several small studies of children with nephrotic syndrome have shown improvements in bone mineral density and bone mineral content with calcium and vitamin D supplementation[17–20].

The National Kidney Foundation and European Society of Pediatric Nephrology provide dosing recommendations for vitamin D supplementation in chronic kidney disease, but the generalizability of these recommendations to those with persistent proteinuria is unknown. [21, 22] The approach to vitamin D supplementation in patients with persistent proteinuria may require modification due to the possible impairment of its intestinal absorption [23] and its excessive elimination in the urine secondary to the loss of its carrier proteins, vitamin D binding protein (VDBP) and albumin. [23–25] Although studies have found that vitamin D supplementation for children with nephrotic syndrome leads to increased vitamin D levels, [18, 26–28] the vast majority of the participants in these studies had steroid responsive nephrotic syndrome and were in remission or went into remission during the supplementation period. Additionally, a recent study found that children with glomerular disease require higher doses of vitamin D supplementation and take longer to achieve optimal levels compared to children with CKD of non-glomerular etiology. [29]

The efficacy and safety of vitamin D supplementation in children with persistent proteinuria who have vitamin D deficiency is unclear. A previous study of seven adult patients with active nephrotic syndrome showed an elevation of serum 25OHD for up to 72 hours after oral administration without any changes in serum or urinary calcium or magnesium, but an increase in serum phosphorus. Despite a rapid increase in 25OHD concentration, intestinal absorption was slower and elimination faster in those with nephrotic syndrome compared to controls, suggesting that the metabolism of vitamin D supplementation may be altered in this setting.[30] Accordingly, our study sought to determine if the administration of vitamin D supplementation over 12 weeks in children and young adults with glomerular disease and persistent proteinuria would lead to an increase in serum 25OHD concentrations without adversely affecting other mineral metabolites. Our primary outcome was change in vitamin D concentration after 12 weeks of supplementation.

Methods

Participants were recruited from four pediatric nephrology centers: Children’s Hospital of Philadelphia (CHOP), Nemours Alfred I. DuPont Hospital for Children, Cohen Children’s Medical Center and Johns Hopkins Medical Center. Study procedures for participants recruited from Nemours were conducted at CHOP. The Institutional Review Boards at CHOP, Cohen Children’s Medical Center and Johns Hopkins Medical Center approved this study. IRB oversight for participants from Nemours was provided by the CHOP IRB through a cooperative agreement. Informed consent was obtained from all participants. The study was performed in accordance with the ethical standards as laid down in the 1964 Declaration of Helsinki and its later amendments or comparable ethical standards. Inclusion criteria included: age ≥ 5 years, a primary glomerular disorder (such as steroid-resistant nephrotic syndrome, minimal change nephropathy, focal segmental glomerulosclerosis (FSGS), membranous nephropathy, membranoproliferative glomerulonephritis, and IgA nephropathy) without systemic inflammatory disorders (i.e. lupus, vasculitis), serum 25OHD concentration <30 ng/mL and urine protein-to-creatinine ratio (UPCR) ≥0.5 at the screening visit. Exclusion criteria included an estimated glomerular filtration rate (eGFR) <30 ml/min/1.73m², serum phosphorus >5.5 mg/dl, hypercalcemia, history of kidney stones, chronic medical conditions or medications unrelated to the glomerular disease that could impact vitamin D status, and pregnancy. eGFR was calculated using the “bedside Schwartz” equation [31] in participants <18 years of age and by the CKD-EPI equation in participants ≥18 years of age. [32]

Participants received 12 weeks of cholecalciferol supplementation according to their baseline 25OHD concentration, as recommended by the National Kidney Foundation.[33] The dosages were modified from National Kidney Foundation recommendations for simplicity. If the 25OHD was <20 ng/mL, they received a daily dose of cholecalciferol of 4000 IU for 12 weeks, and if the 25OHD was 20 to <30 ng/mL they received a daily cholecalciferol dose of 2000 IU for 12 weeks. Participants were evaluated at 6 ± 2 and 12 ± 2 weeks. Participants were provided with an 8-week supply of cholecalciferol at the screening visit and at the week 6 visit. Adherence to vitamin D supplementation was calculated based on pill counts. The participants brought their vitamin D supplement bottle(s) to the week 6 and week 12 visits, and the study staff counted the number of pills remaining from what was prescribed to estimate percent adherence. Adherence was defined as taking ≥ 85% of the prescribed pills. Dietary intake of vitamin D, including from multi-vitamins, was estimated from the average of three 24-hour diet recalls by phone interview conducted at baseline and week 12 (Nutrition Coordinating Center, University of Minnesota Nutrition Data System, Minneapolis, MN).[34]

Serum 25OHD was measured in real time by high-performance liquid chromatography-tandem mass spectrometry (HPLC-MS/MS) in the CHOP clinical laboratory and represents the sum of 25(OH)D₂ and 25(OH)D₃. This 25(OH)D assay gave a linear response from 1.3-135 ng/mL for both 25(OH)D₂ and 25(OH)D₃, and the limit of quantitation was 1.3 ng/mL. The interassay CVs were 7.3-10.0% and 4.2-4.9% for 25(OH)D₂ and 25(OH)D₃, respectively. Values reported as <3 ng/mL were set to 3 ng/mL for analysis.

Safety was assessed with measures of serum 25OHD concentration, serum calcium, whole blood ionized calcium, serum phosphorus and urine calcium excretion at all visits. Values of urine calcium <1.0 mg/dl were set to 1 for analysis. Actionable values and the safety protocol are outlined in the Supplemental Material. All enrolled participants were followed until the end of the study even if cholecalciferol supplementation was discontinued. Serum creatinine, albumin and intact parathyroid hormone (PTH) were also measured in real time by enzymatic assay, colorimetric assay (bromocresol green) and chemiluminescent immunoassay, respectively. We measured vitamin D and related metabolites at the baseline, week 6 and week 12 visits for all participants. Serum samples were stored at −80°C and analyzed in batch at the conclusion of the study at the University of Washington for VDBP, 25OHD, 1,25-dihydroxyvitamin D (1,25OH₂D), 24,25-dihdyroxyvitamin D (24,25OH₂D), and fibroblast growth factor 23 (FGF23). Serum 25OHD, 1,25OH₂D, and 24,25OH₂D were measured by immunoenrichment and HPLC-MS/MS. The full method of extraction and analysis for 25(OH)D has been described previously.[35] The calibration for the measurement of 25(OH)D was verified using the NIST Standard Reference Material 972 with accuracy of 91-95% for 25(OH)D₃ and 100-116% for 25(OH)D₂. The method has lower limits of quantification (20% CV) of 0.5 ng/mL, 1.0 ng/mL, and 5.0 ng/mL for 24,25(OH)₂D₃, 25(OH)D₃, and 25(OH)D₂, respectively. Inter-assay imprecision is 8.58% at 1.5 ng/mL for 24,25(OH)₂D₃, 4.40% at 10.4 ng/mL for 25(OH)D₃, and 4.35% at 9.7 ng/mL for 25(OH)D₂. Free 25(OH)D is a function of the isoform of VDBP, and combinations of two single-nucleotide polymorphisms produce three major DBP isoforms (Gc1f, Gc1s, and Gc2) that differ in their geographical and racial distribution and their affinities for vitamin D metabolites and circulating concentrations.[36–38] We measured VDBP concentration and identified the VDBP isoforms present using LC-MS/MS. [37] Based on the VDBP and VDBP isoforms for each participant, we also calculated the free and bioavailable vitamin D concentrations.[39]

Statistical Analysis

Analyses were performed using R 4.1 and STATA 16. For our analyses, eGFR was calculated using the Chronic Disease in Children (CKiD) Under 25 GFR estimating equations.[40]. Changes in measures of vitamin D and mineral metabolism were assessed using the Wilcoxon signed-rank test. The values for participants who discontinued the cholecalciferol at the week 6 visits were not included in the reporting of data for week 12. Generalized estimating equation (GEE) regression analysis adjusting for age, sex, season of supplementation (May through October versus November through April), eGFR, albumin, UPCR, VDBP and VDBP isoform (Gc1f present versus not) was used to determine if serum 25OHD concentrations increased, after accounting for confounding factors. Visits from the same participants were considered to have exchangeable covariance. For the GEE model, 25OHD values were re-scaled to each patient’s baseline value, and the week 12 visit data for two participants was excluded due to discontinuation of vitamin D supplementation at week 6. An additional GEE regression analysis was also performed using the log-transformed 25OHD values as the outcome to estimate mean ratios of 25OHD levels across visits and for other risk factors. Sequential levels of adjustment were performed for this model to evaluate the associations of covariates and to assess for any signs of model overfitting given the large number of covariates and relatively small number of observations. P values <0.05 were considered statistically significant.

Results

Of 35 screening visits among 31 participants, eighteen met criteria to receive cholecalciferol supplementation. Table 1 shows baseline demographic and clinical data. Median age was 13 years (IQR: 7, 16). The most common diagnosis was FSGS. Median eGFR was 90.3 ml/min/1.73m² (IQR: 70.4, 110.3) and most of the participants were taking prednisone. Fifteen participants at each study visit returned pill bottles to perform pill counts for adherence. At the week 6 visit 73% of participants were adherent, as defined by taking at least 85% of the prescribed doses, and at the week 12 visit 60% were adherent. All participants who returned pill bottles took >50% of the prescribed pills. The median daily vitamin D intake from diet and other non-prescribed (eg multi-vitamin) sources was 222 (IQR: 99-466) units at the baseline visit and 186 (IQR: 61, 348) units at week 12.

Table 1:

Baseline characteristics of participants

	N (%) or Median (IQR)
Age (years)	13 (7, 16)
Male sex	9 (50%)
Black race	8 (44%)
Ethnicity
Does not know	1 (6%)
Not Hispanic or Latino	15 (83%)
Hispanic or Latino	2 (11%)
Haplotype
Gc1f/Gc1f	7 (39%)
Gc1f/Gc1s	5 (28%)
Gc2/Gc1f	3 (17%)
Gc2/Gc1s	1 (6%)
Gc1s/Gc1s	2 (11%)
Supplemented in winter (November to April)	8 (44%)
Etiology of disease
Focal segmental glomerulosclerosis	13 (72%)
Other (Minimal change nephropathy, IgA nephropathy, membranoproliferative glomerulonephritis)	5 (28%)
Estimated glomerular filtration rate (ml/min/1.73m2)	90.3 (70.4, 110.3)
Taking an ace inhibitor or angiotensin receptor blocker	7 (39%)
Taking prednisone	14 (78%)
Taking calcineurin inhibitor	9 (50%)

The Figure and Tables 2 and 3 show the distributions of measures of vitamin D and mineral metabolism at the baseline and follow-up study visits. At baseline, the median 25OHD concentration was 12.8 ng/mL; 16 of the 18 participants had a serum 25OHD level <20 ng/mL. The median 25OHD concentration increased to 27.8 ng/mL by week 6 (p<0.001) without further significant increase at week 12 (median 28.4 ng/mL). At week 12, 31% (5 participants) had a level ≥30 ng/mL, and 19% (3 participants) had a level <10 ng/mL. 24,25OH₂D concentrations increased significantly at week 6 and further increased at week 12 (p<0.001 and p=0.002, respectively). With supplementation, there was a significant increase in bioavailable and free 25(OH)D from the baseline visit to the week 6 and week 12 visits. The ratio of 24,25OH₂D:25OHD increased from baseline to 12 weeks. There were no significant changes in ionized or total calcium, phosphorus, 1,25OH₂D, FGF23, or VDBP.

Figure 1.

Distributions of measures of vitamin D and mineral metabolism at baseline and follow-up study visits

Table 2.

Measures of vitamin D metabolites and vitamin D binding protein at the baseline and follow-up visits

Factor	Baseline median (IQR)	6 weeks median (IQR)	12 weeks median (IQR)	P value week 6 vs baseline	P value week 12 vs baseline	P value week 12 vs week 6
N	18	18	16
25OHD ng/mL (CHOP)	12.8 (9.3-18.9)	27.8 (20.5-36.0)	28.4 (22.3-32.0)	<0.001	<0.001	0.105
25OHD ng/mL (UW)	12.9 (9.4-18.9)	25.8 (20.0-30.7)	28.1 (19.9-35.5)	<0.001	<0.001	0.018
25OHD (CHOP)				<0.001	0.002	0.824
<10 ng/mL	6 (33%)	2 (11%)	3 (19%)
10 to <20 ng/mL	10 (56%)	1 (6%)	0 (0%)
20 to <30 ng/mL	2 (11%)	7 (39%)	8 (50%)
≥30 ng/mL	0 (0%)	8 (44%)	5 (31%)
25OHD (UW)				0.001	0.002	0.233
<10 ng/mL	5 (28%)	2 (11%)	1 (6%)
10 to <20 ng/mL	11 (61%)	2 (11%)	3 (19%)
20 to <30 ng/mL	1 (6%)	9 (50%)	7 (44%)
≥30 ng/mL	1 (6%)	5 (28%)	5 (31%)
1,25OH2D pg/mL	61.0 (48.1-75.9)	61.8 (47.1-82.6)	66.2 (46.8-78.9)	0.671	0.562	0.144
24,25OH2D ng/mL	0.5 (0.4-1.1)	1.6 (1.1-2.6)	1.8 (1.4-3.0)	<0.001	<0.001	0.002
24,25OH2D:25OHD ratio	0.05 (0.03-0.06)	0.06 (0.05-0.08)	0.07 (0.06-0.13)	<0.001	<0.001	0.669
Vitamin D binding protein ug/mL	246.6 (223.7-259.6)	239.0 (218.8-278.4)	237.2 (215.4-275.8)	0.393	0.211	0.464
Free 25OHD pg/mL	3.4 (1.9-5.7)	6.3 (4.5-8.6)	6.0 (4.5-8.8)	<0.001	<0.001	0.074
Bioavailable 25OHD ng/mL	1.0 (0.4-1.2)	2.1 (1.4-3.2)	2.1 (1.3-3.3)	<0.001	<0.001	0.193

Paired Wilcoxon Signed Rank test used for all P-values. Missing values were dropped from analyses. Although some tests in Tables 2 and 3 are dependent, treating all 55 tests as independent yields a conservative Bonferroni-adjusted significance level of about 0.0009.

Table 3:

Measures of mineral metabolism and clinical factors at the baseline and follow-up visits

Factor	Baseline median (IQR)	6 weeks median (IQR)	12 weeks median (IQR)	P value week 6 vs baseline	P value week 12 vs baseline	P value week 12 vs week 6
N	18	18	16
Ionized calcium mmol/L	1.17 (1.15-1.21)	1.20 (1.17-1.21)	1.17 (1.13-1.19)	0.193	0.599	0.276
Serum total calcium mg/dL	8.9 (8.6-9.1)	9.2 (8.6-9.7)	8.9 (8.6-9.3)	0.082	0.410	0.624
Serum phosphorus mg/dL	4.2 (3.9-4.9)	4.4 (4.2-4.9)	4.4 (3.9-5.0)	0.532	0.877	0.660
Intact parathyroid hormone pg/mL	45.7 (27.6-66.7)	----	37.1 (32.1-54.4)	----	0.706	----
Fibroblast growth factor 23 pg/mL	63.3 (51.6-91.4)	61.9 (50.9-103.1)	55.3 (47.6-99.7)	0.523	0.252	0.074
Serum albumin mg/dL	3.4 (2.4-3.7)	3.6 (3.2-4.1)	3.3 (2.8-4.0)	0.148	0.820	0.525
eGFR ml/min/1.73m2	90.3 (70.4-110.3)	80.4 (73.2-93.5)	82.8 (63.4-112.4)	0.468	0.782	0.298
Urine protein:creatinine ratio	2.8 (0.9-7.0)	1.8 (0.7-7.1)	2.8 (0.9-5.9)	0.024	0.900	0.404
Urine calcium:creatinine ratio	0.02 (0.01-0.08)	0.06 (0.02-0.11)	0.02 (0.02-0.05)	0.369	0.977	0.044

Paired Wilcoxon Signed Rank test used for all P-values. Missing values dropped from analyses. Although some tests in Tables 2 and 3 are dependent, treating all 55 tests as independent yields a conservative Bonferroni-adjusted significance level of about 0.0009.

Cholecalciferol supplementation was discontinued in two participants at week 6: one for an ionized calcium of 1.36 mmol/L with a 25OHD concentration of 43.4 ng/mL (baseline 25OHD concentration was 11.4 ng/mL) and one for a phosphorus of 5.2 mg/dL with an associated 25OHD level of 52.2 ng/mL (baseline 25OHD concentration was 9.3 ng/mL). Both participants experienced subsequent declines in 25OHD concentration of more than 20 ng/mL after stopping the supplementation. The participant with mild hypercalcemia experienced a marked improvement in serum albumin between baseline and week 6 (2.7 to 4.4 g/dL) despite no change in proteinuria. The participant with mild hyperphosphatemia experienced remission from proteinuria and an associated improvement in serum albumin from baseline to week 6 (2.1 to 4.1 g/dL), although by week 12 had relapsed. Based on pill count data at the week 6 visit, we suspect that this participant over-consumed cholecalciferol. There were no other participants who experienced remission of proteinuria during the study period. Another participant experienced a decrease in 25OHD concentration from week 6 to week 12 of 25 ng/mL in the setting of acute worsening of proteinuria during that time (UPCR 2.6 to 6.0) and a decline in serum albumin from 3.9 to 2.5 g/dL. None of the participants developed hypercalciuria or vitamin D excess (serum 25OHD >80 ng/mL).

In multivariable GEE analysis, adjusting for age, sex, season of supplementation, VDBP isoform, eGFR, albumin, UPCR, and VDBP, 25OHD concentrations increased by 102% (95% CI: 64, 141) from baseline to week 6 and 96% (95% CI: 51, 140) from baseline to week 12 (Table 4) for the patients receiving cholecalciferol therapy. Results were qualitatively similar using the log-transformed 25OHD models, whether there were fewer or more adjustment covariates included in the model. In the final adjusted model, older age, higher eGFR, lower albumin, higher UPCR, having a Gc1f haplotype, and lower VDBP were also significantly associated with lower 25OHD levels during the study period. (Supplemental Table 1)

Table 4:

Mutually adjusted Generalized Estimating Equation to determine percent change in 25-hydroxyvitamin D concentration over the study period

Variable	Mean Percentage Point Change in Vitamin D (95% CI)	P value
Visit
Week 6 Visit vs baseline	102 (64, 141)	<0.001
Week 12 Visit vs baseline	96 (51, 140)	<0.001

Adjusted for baseline age, sex, supplementation season, Gc1f diplotype, eGFR, albumin, urine protein:cr and vitamin D binding protein. Exchangeable covariance assumed between visits from the same participant. Visits after discontinuation of calciferol supplementation were excluded from the analysis (2 visits).

Discussion

Vitamin D deficiency is highly prevalent in children with nephrotic syndrome [13] and may impact disease progression and complications.[7–10] Although there are other studies describing effective treatment of low vitamin D levels in children with nephrotic syndrome, these studies almost exclusively evaluated children who had steroid sensitive nephrotic syndrome.[18, 26] This study specifically evaluated cholecalciferol supplementation for vitamin D deficient children with persistent proteinuria secondary to glomerular disease. At the end of 12 weeks of supplementation, 25OHD concentrations significantly increased from baseline, without any serious adverse events and with very minimal changes in associated mineral metabolism. Additionally, there was a significant increase in bioavailable and free 25(OH)D. Vitamin D supplementation was discontinued in two participants after 6 weeks of supplementation for mild hypercalcemia and hyperphosphatemia respectively, in conjunction with 25OHD concentrations >30 ng/mL. Both participants experienced a concurrent improvement in serum albumin from the hypoalbuminemic to normal range, thereby highlighting the influence that serum albumin has on vitamin D status and associated labs of mineral metabolism. Both participants also subsequently experienced dramatic reductions in 25OHD concentrations to deficient levels with discontinuation of supplementation and ongoing or relapse of proteinuria, highlighting the strong influence of proteinuria on vitamin D levels. These examples also illustrate the need to re-examine vitamin D dosing when there is a marked change in either serum albumin or urine protein levels.

Despite supplementing vitamin D at doses similar to established CKD guidelines[33], most children failed to achieve a level ≥30 ng/mL after 12 weeks of supplementation, suggesting that our dosage algorithm may be inadequate. Non-adherence could explain the sub-optimal vitamin D levels achieved in many participants. It is unclear whether target vitamin D levels would be achieved with a more prolonged supplementation course since there was not a further significant increase in 25OHD concentration from week 6 to week 12.

In our study, despite the significant increase in 25OHD concentrations, there was not a significant decrease in PTH levels. This is consistent with prior literature on vitamin D supplementation in healthy adults and adults with CKD in which neither group demonstrated a change in PTH or FGF23 in response to increasing 25OHD concentrations.[41] In contrast, a study of pediatric patients with CKD, 17% of whom had glomerular disease, found that 6 weeks after supplementation with 60,000 IU of cholecalciferol divided over three days, the median PTH significantly decreased.[42] Our study’s lack of a change in PTH may be because, despite low 25OHD concentrations at baseline, the median PTH level was well below the threshold for hyperparathyroidism, as has been shown in other studies in children with nephrotic syndrome[43].

This study is strengthened by the evaluation of changes in 24,25OH₂D, the product of CYP24A1-mediated catabolism of 25OHD that is stimulated by 1,25OH₂D. The ratio of 24,25OH₂D:25OHD significantly increased from baseline suggesting a change in vitamin D activity in addition to and beyond the change in 25OHD level.[44] 25OHD is not the active form of vitamin D and therefore, may not fully represent vitamin D receptor activity by 1,25OH₂D. 24,25OH₂D concentrations may reflect the downstream activity of 25OHD and thereby serve as a more meaningful indicator of tissue level vitamin D activity. This hypothesis is supported by a study of community dwelling adults that showed that 25OHD concentrations did not correlate with hip fracture, but higher 24,25OH₂D concentrations and higher 24,25OH₂D:25OHD ratios both correlated with a lower risk of hip fracture.[45] Lower 24,25OH₂D:25OHD ratio was also found to be associated with all-cause mortality in a large cohort of adults with CKD.[46] Additonally, unlike 25OHD, the 24,25OH₂D:25OHD metabolite ratio is independent of VDBP, [47] which may be affected in proteinuric states due to urinary losses.

To further evaluate the importance of vitamin D supplementation in the setting of nephrotic syndrome, we calculated the free and bioavailable 25(OH)D concentrations. Despite some concern that low vitamin D levels in the setting of nephrotic syndrome do not reflect a true vitamin D deficiency, we found low levels of free and bioavailable vitamin D levels in the participants at baseline suggesting a true deficiency. [41, 48–50] The free and bioavailable vitamin D levels significantly increased with supplementation. The relevance of free and bioavailable vitamin D to related outcomes has been previously reported.[12, 48, 50, 51] Contrary to our findings, a study by Banerjee et al did not find lower free 25(OH)D concentrations in participants with steroid sensitive nephrotic syndrome compared to controls , despite finding significantly lower levels of total 25(OH)D. This may be due to the different patient populations studied. The difference in findings highlight the importance of duration of proteinuria when supplementing with vitamin D, and the lack of generalizability between patients with steroid sensitive nephrotic syndrome and those with persistent proteinuria.

Limitations of this study include the small sample size, although we were able to detect significant correlates of response to vitamin D therapy. We also do not have a control group of participants with persistent proteinuria who were not supplemented with vitamin D over this time. Lastly, since we did not evaluate a longer course of therapy or analyze vitamin D concentration changes in relation to the supplementation dose, the best prescription to achieve sufficient vitamin D levels in this population is still not clear.

This is the first study evaluating the efficacy and safety of vitamin D supplementation in children with glomerular disease and persistent proteinuria. Altough vitamin D levels improved after 12 weeks of supplementation in the setting of ongoing proteinuria, vitamin D levels remained <20 ng/ml in almost 20% of the cohort. The NKF suggested dosing regimen for CKD did not reliably achieve goal vitamin D levels ≥30 ng/mL in persistently proteinuric children with glomerular disease. The ideal vitamin D regimen to achieve higher rates of vitamin D sufficiency in this population still needs to be determined.