Vitamin d supplementation and bone health in children with nephrotic syndrome: A systematic review and meta-analysis

Rema G Caronan-Parallag; Tristan Marvin Z. Uy; Francisco E Anacleto; Eric Emmanuel T. Aragon; Lourdes Paula Real Resontoc

doi:10.4103/ajpn.ajpn_35_20

ORIGINAL ARTICLE

Year : 2021 | Volume : 4 | Issue : 1 | Page : 6-13

Vitamin d supplementation and bone health in children with nephrotic syndrome: A systematic review and meta-analysis

Rema G Caronan-Parallag¹, Tristan Marvin Z. Uy², Francisco E Anacleto³, Eric Emmanuel T. Aragon¹, Lourdes Paula Real Resontoc¹
¹ Department of Pediatrics, Division of Pediatric Nephrology, University of the Philippines - Philippine General Hospital, Manila, Philippines
² Department of Pediatrics, Division of Pediatric Nephrology, University of the Philippines - Philippine General Hospital; Pediatric Infectious Disease and Tropical Medicine Department, San Lazaro Hospital, Manila, Philippines
³ Department of Pediatrics, Division of Pediatric Nephrology, University of the Philippines - Philippine General Hospital; Department of Physiology, College of Medicine, University of the Philippines, Manila, Philippines

Date of Submission	27-Oct-2020
Date of Decision	31-Jan-2021
Date of Acceptance	13-Apr-2021
Date of Web Publication	30-Jun-2021

Correspondence Address:
Lourdes Paula Real Resontoc
Department of Pediatrics, Division of Pediatric Nephrology, University of the Philippines-Philippine General Hospital, Taft Avenue, Manila 1000
Philippines

Source of Support: None, Conflict of Interest: None

DOI: 10.4103/ajpn.ajpn_35_20

Abstract

Background: While steroids have been the standard treatment in nephrotic syndrome (NS), they are known to deleteriously affect bone mineralization. Objectives: The objectives were to determine the efficacy and safety of vitamin D supplementation among children with NS on steroid therapy. Methods: We searched databases, scanned reference lists, and contacted trial investigators. Two reviewers collected and graded randomized controlled trials comparing oral vitamin D3 with placebo or no intervention in terms of bone mineral content/density (BMC/BMD), serum markers, and adverse events in pediatric NS. Results: We included 4 trials (164 subjects) with a collectively high risk of performance and/or detection bias. Although the vitamin D group had significantly more positive absolute change-from-baseline BMC than controls (mean difference 1.15; 95% CI 0.07-2-22;I 62%), the two trials were heterogeneous and included data. Moreover, in terms of other outcome measures, we did not find sufficient evidence of benefit from treatment. One study reported significantly improved parathyroid hormone levels but also a higher risk of hypercalciuria with vitamin D use. No nephrocalcinosis was reported. Subgroup analysis of first-episode NS revealed significantly more improved BMD with supplementation. Conclusion: Available evidence was low-quality and insufficient to recommend vitamin D supplementation in pediatric NS, although there may be benefit in first-episode NS. In contrast, vitamin D administration may be associated with hypercalciuria.

Keywords: Child, nephrotic syndrome, steroids, Vitamin D

How to cite this article:
Caronan-Parallag RG, Z. Uy TM, Anacleto FE, T. Aragon EE, Real Resontoc LP. Vitamin d supplementation and bone health in children with nephrotic syndrome: A systematic review and meta-analysis. Asian J Pediatr Nephrol 2021;4:6-13

How to cite this URL:
Caronan-Parallag RG, Z. Uy TM, Anacleto FE, T. Aragon EE, Real Resontoc LP. Vitamin d supplementation and bone health in children with nephrotic syndrome: A systematic review and meta-analysis. Asian J Pediatr Nephrol [serial online] 2021 [cited 2023 May 28];4:6-13. Available from: https://www.ajpn-online.org/text.asp?2021/4/1/6/320188

Introduction

Nephrotic syndrome (NS), a common glomerular disease in children, is characterized by nephrotic-range proteinuria, hypoalbuminemia, and edema.^[1],[2],[3] While corticosteroids have been the standard treatment, their long-term use has also been associated with adverse effects, including osteoporosis.^[1],[2],[3] They have been known to deleteriously affect bone mineral acquisition by promoting osteoblastic apoptosis, activating osteoclasts, and centrally inhibiting growth hormone secretion.^[1],[2],[3]

Skeletal mineral acquisition during childhood may be measured through bone mineral content (BMC) or bone mineral density (BMD).^[4],[5],[6] Since BMD (BMC divided by bone area) may be confounded by differing bone sizes, it has not been considered a measurement of true volumetric bone density.^[4],[5],[6] Therefore, the BMC, accounting for bone girth and amount of deposited mineral, has been deemed the more accurate reflection of bone mass.^[4],[5],[6] Despite established effects of steroids on bone health, there has been no consensus on supplementation with vitamin D in NS, and various doses of vitamin D have been used in trials.^{[7],[8],[9],[10],[11],[12],[13]} In view of this growing body of literature and perceived need for unified recommendations on vitamin D supplementation, we conducted a review of its benefits and safety in pediatric NS.

Methods

We followed the Preferred Reporting Items for Systematic Review and Meta-analysis (PRISMA) guidelines throughout the design, implementation, analysis, and reporting of our meta-analysis.^[14]

Inclusion criteria

Types of studies

We included randomized controlled trials (RCTs) published in English. Observational studies, nonrandomized trials, controlled before-after studies, interrupted time series, repeated measures study, case reports, and case series were excluded.

Types of participants

RCTs involving children below 19 years of age with steroid-sensitive NS, and RCTs involving adults and children, where data for children could be extracted separately.

Types of interventions

We included trials that compared vitamin D (or a vitamin D analog) in any form (i.e., tablet, injectable solution), dosage, or duration with either of the following: a placebo, an active control, or no intervention. No restrictions were placed on the presence of co-interventions or steroid regimen used for therapy of NS. Studies comparing different dosages of vitamin D without a negative control were excluded.

Outcome measures

Our primary outcome was the absolute change-from-baseline BMC or BMD, measured via dual-energy X-ray absorptiometry (DXA) scan at the lumbar spine. Variations of this outcome included postintervention scores and percentage (%) change in scores, both of which we deemed less preferable due to lack of precision and statistical power.^[15],[16] In all these versions of the outcome, a more positive result represented benefit.

Our secondary outcomes included the absolute change from baseline (or variations thereof) in serum markers of bone health, which were any of the following: calcium (Ca), phosphate (P), alkaline phosphatase (ALP), vitamin D, or parathyroid hormone (PTH) levels. A more negative change in ALP or PTH or a more positive change in calcium, phosphate, or vitamin D levels signified benefit. Adverse events constituted another secondary outcome.

Search methods

We systematically searched the following databases from August 1, 2020, to October 2, 2020, for published studies and ongoing trials: Cochrane Central Register of Controlled Trials (CENTRAL), MEDLINE (via PubMed 1966 to 2020), World Health Organization International Clinical Trials Registry Platform (WHO ICTRP), Acta Medica, and Health Research and Development Information Network. We used medical subject headings and text word terms and restricted results to RCTs, clinical trials, reviews, systematic reviews, and meta-analysis. “Vitamin D”, ”nephrotic syndrome,” and “children” composed our main search terms [Supplementary Table 1]. To find other studies, we performed a search in Google Scholar and hand-searched reference lists.

Data collection and analysis

Selection of studies

Two authors (RCP, TU) independently screened titles and abstracts and reviewed the full text of potentially eligible studies. A third author (LR) was consulted to reach a consensus in cases of disagreement. We detailed the process of selection and exclusion in a PRISMA flow diagram and a “Characteristics of excluded studies” [Supplementary Table 1] and [Supplementary Table 2].^[14]

Data extraction and management

One author (RCP) completed a standardized data extraction form for study characteristics, while another (TU) checked for accuracy and completeness. Each of the two authors extracted outcome data, compared notes, and resolved disagreements, if any, by consulting another author (LR). After resolving disagreements and removing duplications, a singular outcome data set was then transferred to Review Manager 5.4 (RevMan).^[17] We attempted to contact study authors for missing data.

Assessment of risk of bias in included studies

As per the Cochrane Handbook for Systematic Reviews of Interventions,^[15] two authors separately graded the risk of bias as “high, low, or unclear” in five domains (sequence generation, allocation concealment, blinding, incomplete outcome data, and selective reporting). Disagreements were resolved through discussion with another author to reach a consensus.

Measures of treatment effect

We used RevMan to calculate treatment effects,^[17] expressed as mean differences (MD) for continuous outcomes and odds ratios (OR) for dichotomous outcomes. A 95% confidence interval (CI) was applied. When studies reported medians and interquartile ranges (IQR), the results were converted to means and standard deviations (SDs) using methods described by Wan et al. and based on the Cochrane Handbook.^[15],[18]

Dealing with missing data

Trial investigators were contacted for missing data. Where attempts were unsuccessful, we derived missing data (i.e., standard deviations) from other available statistics (i.e., P values) or imputed with the average of data from other trials, as outlined in the Cochrane Handbook.^[15]

Assessment of heterogeneity

The I² statistic was used to assess heterogeneity among studies. Value ≥ 50% represented significant heterogeneity.

Assessment of reporting biases

We planned to do a funnel plot analysis if we were able to pool more than 10 trials.

Data synthesis

Where more than one study provided usable data for a particular comparison, we used RevMan to pool continuous data by “Inverse-Variance method” and dichotomous data by “Mantel-Haenszel method”, applying a random-effects model.^[17]

Evidence quality

The five GRADE considerations (study limitations, consistency of effect, imprecision, indirectness, and publication bias) were used to judge evidence quality, as prescribed in the Cochrane Handbook.^[15]

Subgroup analysis and investigation of heterogeneity

In anticipation of a heterogeneous study population in terms of NS classification, we planned to conduct subgroup analyses for first-episode NS, relapse, and each subtype of NS if more than one study provided usable data for a single comparison. We also planned to do subgroup analysis for studies with similar timing of outcome assessment or length of follow-up that did not vary by more than 6 weeks.

Sensitivity analysis

Aside from the aforementioned subgroup analyses, we planned to perform sensitivity analyses by excluding trials with a high risk of bias and studies with missing data and by imputing data from different sources. We also planned to explore the consistency of treatment effect across different outcome measures (i.e. absolute change from baseline, postintervention values, % change).

Results

Results of the search

We found a total of 292 records from our search of the following databases [Figure 1]: CENTRAL (36 records), MEDLINE (80 records), WHO ICTRP (29 records), and Google Scholar (147 records). Local databases and hand-searching of reference lists yielded no additional records. After removal of 183 duplicates, 109 titles and abstracts were screened, 16 of which then underwent full-text review. Finally, we included four studies in the systematic review [Supplementary Table 2] and Supplementary Table 3],^{[7],[8],[19],[20]} after excluding twelve articles.M^{[13],[21],[22],[23],[24],[25],[26],[27],[28],[29],[30],[31]} We attempted to contact the authors of Choudhary 2014 for missing data, but were unsuccessful.

Figure 1: Study flow diagram

Click here to view

Included studies

All included studies were small-scale RCTs (sample size: 40–43) with a total of 164 participants [Table 1]. Three trials were conducted in India^{[7],[19],[20]}, and one in Turkey.[8] While all studies used oral vitamin D3 (cholecalciferol), they applied a variety of dosages (400–8500 IU/day) and regimens (single dose, daily, or once weekly for 4–12 weeks). All trials also supplemented the treatment group with calcium at a range of doses (200–1000 mg/day) and durations (8–12 weeks). Moreover, different protocols were followed for NS therapy across studies. One study[8] used weight-based steroid dosing for 8 weeks, while another^[7] implemented the body-surface-area method for 12 weeks. Two studies^[19],[20] did not adequately describe the steroid course.

Table 1: Summary of key characteristics of the included studies

Click here to view

All studies reported at least one of the primary outcomes (i.e., BMC, BMD) of this review and used a DXA scan of the lumbar spine for measurement. However, trials differed widely in the timing of outcome assessment (i.e., length of follow-up): 8 weeks^[8], 12 weeks^[7],[20], and 24 weeks.[19] Furthermore, one study[19] reported medians instead of means, while another[7] had missing standard deviations. One study[20] did not report percentage changes. Meanwhile, all studies reported at least one serum marker of bone health. Although Banerjee et al. performed serum chemistries at 6 weeks and 24 weeks, only data from the earlier time point were included in our analysis due to its proximity to time points in other trials.^[19] Only two studies determined presence of an adverse event: hypercalciuria expressed in urinary calcium levels in 1 trial[8] and calcium–creatinine (Ca/Cr) ratios in the other.[19] The latter study also performed renal ultrasounds to establish the presence of nephrocalcinosis among subjects with hypercalciuria.

Risk of bias

The studies had a low or unclear risk of bias in most domains but were collectively at high risk of performance and detection bias [Supplementary Figure S1] and [Supplementary Figure S2].

Allocation

Majority of the trials used a computer-generated sequence for randomization. Only one study[8] did not specify the method applied and was deemed to have an unclear risk of selection bias. In contrast, one study[20] concealed allocation through opaque envelopes, while the rest were assessed to have an unclear risk for this domain.

Blinding

Since only one study[20] gave the control group placebo, most of the trials could not blind patients and personnel and had a high risk of performance bias. Moreover, the studies had an unclear-to-high risk of detection bias from lack of blinding of outcome assessors.

Incomplete outcome data

One study[19] was judged to have a high risk of attrition bias from an unequal number of dropouts between study groups and absence of sensitivity analysis. This same latter reason conferred an unclear risk to another study.[19]

Selective reporting

The trials reported all predefined outcomes and had a low risk of bias for this domain.

Other potential sources of bias

Apart from the foregoing domains, there were no other sources of bias found.

Effects of intervention

Bone mineral content and bone mineral density

Two studies^[7],[19] supplied data on BMC, but one[19] reported medians instead of means, while[19] did not report SDs. To convert medians to means, we utilized an IQR-based formula proposed by Wan et al., similar to an estimator in the Cochrane handbook that provided more accurate estimates for small sample sizes and that had performed adequately in 1000 computer-generated simulations.^[14],[18] Meanwhile, missing SDs in Choudhary 2014 were imputed from the average of SDs in Banerjee 2017.^[7],[19] After pooling results (84 patients), we found a significantly more positive absolute change in BMC from baseline in the treatment group compared to the control group (MD: 1.15 [0.07, 2.22], I² = 62%). However, these results showing benefit were precluded by significant heterogeneity between the two studies and possible risk of bias from the imputation of data [Figure 2]. For sensitivity analysis, we imputed the highest SD, instead of the average. The same treatment effect was detected, but heterogeneity remained, rendering no further support for a treatment benefit [Supplementary Table 4]. Moreover, we did not find evidence of benefit from treatment in terms of postintervention BMC (MD: 0.04 [−3.2, 3.28], I² = 0%) [Figure 3] and BMC% change (MD: 10.92 [−6.64, 28.49], I² = 93%) [Supplementary Figure S3].

Figure 2: Forest plot of comparison: Mean change in bone mineral content from baseline

Click here to view

Figure 3: Forest plot of comparison: Mean bone mineral content after intervention

Click here to view

All four trials provided some data on BMD. However, as with BMC, we had to impute SDs for one[7] and convert medians from another[19] in the manner adapted from Wan et al. and previously described.^[18] Pooling of results (164 subjects) revealed no sufficient evidence of statistically significant difference between treatment and control in terms of either the absolute change in BMD from baseline (MD: 0.02 [0.00, 0.04], I² = 50%) or post-intervention (MD: 0.04 [0.00, 0.08], I² = 43%) [Figure 4] and [Figure 5]. Similarly, no sufficient evidence of difference in percent BMD change (MD: 3.35 [−2.97, 9.67], I² = 92%) was found from pooled results (124 patients) of three studies^[7],[8],[19] [Supplementary Figure S4].

Figure 4: Forest plot of comparison: Mean change in bone mineral density from baseline

Click here to view

Figure 5: Forest plot of comparison: Mean bone mineral density after intervention

Click here to view

We performed multiple sensitivity analyses for BMD by imputing the highest SD, excluding the study with imputed data[7], excluding the study with an outlying length of follow-up[19], and excluding studies with a high risk of bias^[7],[19] [Supplementary Table 4]. Only the last two scenarios showed a significant benefit from treatment, in terms of absolute change-from-baseline BMD, with no significant heterogeneity among trials [Supplementary Table 4]. Different scenarios for postintervention BMD and percent of BMD change also did not consistently reveal statistically significant results or homogeneity among involved trials [Supplementary Table 4].

Serum markers of bone health

All studies presented data on serum Calcium, although one[19] reported medians, while another[7] only the percent changes. Pooled results (83 patients) from three trials^{[8],[19],[20]} revealed no sufficient evidence of difference between treatment and control in terms of absolute change-from-baseline calcium (MD −0.29 [−0.95, 0.37], I² = 0%) or postintervention Ca levels (MD −0.08 [0.28,0.11], I² = 0%) [Supplementary Figure 5] and [Supplementary Figure S6]. Excluding one study with a high risk of bias[19], the sensitivity analysis did not change results [Supplementary Table 4]. Likewise, one study[7] found no sufficient evidence of difference between the study groups in terms of percent change ca (MD: 0.76 [−4.33, 5.85]).

Data on alkaline phosphatase and vitamin D were available from two studies.^[19],[20] Pooled results (83 participants) showed no sufficient evidence of difference in absolute change-from-baseline alkaline phosphatase (MD: 11.46 [−35.16, 58.09], I² = 0%) between the two groups, although there was a significantly higher postintervention ALP in the treatment group, thus favoring controls (MD: 12.59 [5.06, 20.11], I² = 0%) [Supplementary Figure S7] and [Supplementary Figure S8]. We also found no sufficient evidence of benefit from vitamin D in terms of absolute change-from-baseline vitamin D levels (MD: 23.34 [−26.00, 72.68], I² = 92%) or postintervention levels (MD: 26.85 [−26.76, 80.47], I² = 98%) while detecting significant heterogeneity [Supplementary Figure S9] and [[Supplementary Figure S10]. Separately, the two trials showed conflicting results in terms of vitamin D levels.

Two studies^[7],[20] provided data on serum phosphate levels, but the results could not be pooled. In either study, no sufficient evidence of difference was found between the study arms in terms of absolute change-from-baseline phosphate levels [MD: 0.77 [−0.79, 2.33]), postintervention levels (MD: 0.15 [−0.06, 0.36]), or percent change (MD 5.7 [−9.27, 20.67]) [Supplementary Figure S11], [Supplementary Figure S12], [Supplementary Figure S13]. Only one study[19] reported PTH values and found a significant benefit from treatment in terms of absolute change-from-baseline PTH (MD: −9.9 [−14.18, −5.62]) and postintervention levels (MD: −18.2 [−25.22, −11.18]) [Supplementary Figure S14] and [Supplementary Figure S15].

Adverse events

Hypercalciuria was the sole adverse event reported, with data coming from only two studies.^[8],[19] Pooling of results (83 subjects) revealed no sufficient evidence of difference between treatment and control in terms of risk of hypercalciuria (OR: 3.47 [0.1, 125.74], 87%) [Supplementary Figure S16]. However, differences in the definition of hypercalciuria between trials may have contributed to the significant heterogeneity. While one trial also found a higher postintervention Ca/Cr ratio with treatment (MD: 0.14 [0.05, 0.23]), it did not find sufficient evidence of benefit in terms of absolute change-from-baseline Ca/Cr (MD: 0.11 [0.00, 0.22]) [Supplementary Figure S17] and [Supplementary Figure S18].[19] On The other study found no sufficient evidence of benefit from vitamin D in terms of risk of hypercalciuria (OR: 0.64 [0.17, 2.38]), absolute change-from-baseline levels [MD − 0.3 [−1.92, 1.32]), or postintervention levels (MD: 0.6 [−2.11, 0.91]) [Supplementary Figure S17], [Supplementary Figure S19] and [Supplementary Figure S20].[8]

Subgroup analysis

Three studies involved children with first-episode NS. [7, 8, 20] Pooled results (81 subjects) from 2 studies^[7],[20] showed no sufficient evidence of benefit from treatment in terms of absolute change-from-baseline BMD (MD: 0.02 [0.00, 0.05], I² = 59%) or postintervention BMD (MD: 0.04 [−0.03, 0.10], I² = 73%) [Supplementary Figure S21] and [Supplementary Figure S22]. However, imputation with the highest SD during sensitivity analysis revealed a significantly more positive absolute change-from-baseline BMD (MD: 0.03 [0.01, 0.05], I² = 0%) and higher postintervention BMD (MD: 0.05 [0.01, 0.09], I² = 20%) in the treatment group compared to controls, with no significant heterogeneity [Supplementary Figure S23] and [Supplementary Figure S24].

Discussion

Summary of main results

Although pooled results from two studies showed that one outcome measure (change-from-baseline BMC) was significantly more improved with treatment, the two trials were heterogeneous, and synthesis was restricted by missing data. Moreover, sensitivity analyses did not show consistent evidence of benefit in terms of other outcome measures (i.e., postintervention mean, percent change) or BMD.

We found no consistent, sufficient evidence of benefit from vitamin D supplementation in terms of serum markers of bone health. One study found a significant benefit from treatment in terms of more improved PTH levels compared to controls. In contrast, one study found a significantly higher risk of hypercalciuria with vitamin D use. No nephrocalcinosis was reported.

Subgroup analysis among children with first-episode NS showed benefit from the use of vitamin D in terms of improved BMD compared to controls.

Overall completeness and applicability of evidence

Applicability of our results may be constrained by the paucity and small scale of trials found. Heterogeneity in study procedure may further limit applicability where institutional differences exist in NS diagnosis or therapy and vitamin D dosing. The NS subtype and pubertal stage were plausible confounders, while the steroid regimen and calcium supplementation were co-interventions, all of which may affect real-life efficacy of vitamin D. Furthermore, the inconsistent (and possibly inadequate) length of follow-up among the trials may have missed the more clinically significant endpoints (e.g., growth retardation, short stature, fractures, etc.). Not all studies reported adverse events as an outcome, while only one study determined the presence of nephrocalcinosis from vitamin D use. In contrast, all trials were conducted in middle-income countries, thus enhancing the applicability of results to resource-limited settings.

Quality of evidence

The body of evidence was deemed to be of low quality, attributable to the potential risk of bias, inconsistency, imprecision, and indirectness of the studies. The small sizes of the trials contributed to imprecision and inconsistency, while the absence of clinically relevant outcomes added indirectness.

Potential biases in the review process

We adhered to the methods prescribed in the Cochrane handbook to minimize bias.^[15] Moreover, two authors independently conducted the search, extracted data, and pooled results. All reported subgroup analyses were preplanned. Meanwhile, imputations for missing data and secondary statistical derivations were potential sources of bias, which we endeavored to address through multiple sensitivity analyses. Finally, language restriction of the search may have conferred bias.

Agreements and disagreements with other reviews

A previous review, by Bacchetta et al. in 2008, of adjunct therapy (e.g., calcium, vitamin D, growth hormone, etc.) for pediatric NS on long-term steroid therapy could not perform a meta-analysis but suggested a possible beneficial effect of calcium and vitamin D supplementation on bone metabolism.^[26] Two other reviews published in 1984 and 1987 discussed bone metabolism in NS but did not make recommendations on vitamin D supplementation.^[25],[27]

Authors' Conclusions

Implications for practice

We found no sufficient evidence to suggest the use of vitamin D among children with NS on steroid therapy. The available evidence was of low quality and did not show consistent benefit from vitamin D supplementation. There may be some benefit from its use among children with first-episode NS. In contrast, vitamin D supplementation may be associated with a risk of hypercalciuria.

Implications for research

To improve the quality of evidence on the use of vitamin D in pediatric NS, we recommend further large-scale trials with well-defined criteria for the diagnosis of NS; standardized steroid, vitamin D, and calcium supplement regimens; longer duration of follow-up; and clinically relevant outcomes.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.

References

1.	Mushtaq T, Ahmed SF. The impact of corticosteroids on growth and bone health. Arch Dis Child 2002;87:93-6.
2.	Friedman M, Strang LB. Effect of long-term corticosteroids and corticotrophin on the growth of children. Lancet 1966;2:569-72.
3.	Allen DB. Growth suppression by glucocorticoid therapy. Endocrinol Metab Clin North Am 1996;25:699-717.
4.	Heaney RP. Bone mineral content, not bone mineral density, is the correct bone measure for growth studies. Am J Clin Nutr 2003;78:350-1.
5.	Heaney RP. Design considerations for clinical investigations of osteoporosis. In: Marcus R, Feldman D, Dempster DW, Luckey M, Cauley JA, editors. Osteoporosis. 4th ed. Cambridge, MA: Academic Press; 2013. p. 1753-76.
6.	Prentice A, Parsons TJ, Cole TJ. Uncritical use of bone mineral density in absorptiometry may lead to size-related artifacts in the identification of bone mineral determinants. Am J Clin Nutr 1994;60:837-42.
7.	Choudhary S, Agarwal I, Seshadri MS. Calcium and vitamin D for osteoprotection in children with new-onset nephrotic syndrome treated with steroids: A prospective, randomized, controlled, interventional study. Pediatr Nephrol 2014;29:1025-32.
8.	Bak M, Serdaroglu E, Guclu R. Prophylactic calcium and vitamin D treatments in steroid-treated children with nephrotic syndrome. Pediatr Nephrol 2006;21:350-4.
9.	Gulati S, Sharma RK, Gulati K, Singh U, Srivastava A. Longitudinal follow-up of bone mineral density in children with nephrotic syndrome and the role of calcium and vitamin D supplements. Nephrol Dial Transplant 2005;20:1598-603.
10.	Misra M, Pacaud D, Petryk A, Collett-Solberg PF, Kappy M, Drug and Therapeutics Committee of the Lawson Wilkins Pediatric Endocrine Society. Vitamin D deficiency in children and its management: Review of current knowledge and recommendations. Pediatrics 2008;122:398-417.
11.	Calikoglu AS, Davenport ML. Prophylactic vitamin D supplementation. Endocr Dev 2003;6:233-58.
12.	Rathi N, Rathi A. Vitamin D and child health in the 21^st century. Indian Pediatr 2011;48:619-25.
13.	Muske S, Krishnamurthy S, Kamalanathan SK, Rajappa M, Harichandrakumar KT, Sivamurukan P. Effect of two prophylactic bolus vitamin D dosing regimens (1000 IU/day vs. 400 IU/day) on bone mineral content in new-onset and infrequently-relapsing nephrotic syndrome: A randomised clinical trial. Paediatr Int Child Health 2018;38:23-33.
14.	Moher D, Liberati A, Tetzlaff J, Altman DG, PRISMA Group. Preferred reporting items for systematic reviews and meta-analyses: The PRISMA statement. PLoS Med 2009;6:e1000097.
15.	Higgins JPT, Green S (editors). Cochrane handbook for systematic reviews of interventions version 5.1.0. United Kingdom: The Cochrane Collaboration; 2011. Available from www.handbook.cochrane.org. [Last updated Mar 2011].
16.	Vickers AJ. The use of percentage change from baseline as an outcome in a controlled trial is statistically inefficient: A simulation study. BMC Med Res Methodol 2001;1:6.
17.	Nordic Cochrane Centre, The Cochrane Collaboration. Review Manager (RevMan). Ver. 5.4. Copenhagen: Nordic Cochrane Centre, The Cochrane Collaboration; 2014.
18.	Wan X, Wang W, Liu J, Tong T. Estimating the sample mean and standard deviation from the sample size, median, range and/or interquartile range. BMC Med Res Methodol 2014;14:135.
19.	Banerjee S, Basu S, Sen A, Sengupta J. The effect of vitamin D and calcium supplementation in pediatric steroid-sensitive nephrotic syndrome. Pediatr Nephrol 2017;32:2063-70.
20.	Yadav VK, Sharma S, Debata PK, Patel S, Kabi BC, Aggrawal KC. Change in bone mineral density and role of vitamin D and calcium supplementation during treatment of first episode nephrotic syndrome. J Clin Diagn Res 2017;11:C18-21.
21.	Clinical Trials Registry India [Internet]. New Delhi: database publisher (India). 2007 Jun 20 - . Identifier CTRI/2017/08/009339, A study to compare two different doses of vitamin D supplementation in children with nephrotic syndrome; 2017. Available from: www.ctri.nic.in/Clinicaltrials/pmaindet2.php?trialid=14233. [Last cited on 2020 Oct 02].
22.	Iranian Registry of Clinical Trials [Internet]. Tehran: data publisher (Iran). 2008 - . Identifier IRCT20130518013366N11, The effect of vitamin D on nephrotic syndrome in children; 2019. Available from: https://www.irct.ir/trial/37791. [Last cited on 2020 Oct 02].
23.	Clinical Trials Registry India [Internet]. New Delhi: database publisher (India). 2007. Identifier CTRI/2016/10/007405, Comparison of the efficacy of two dosing regimens of Vitamin D for bone protection in children with difficult nephrotic syndrome; 2016. Available from: www.ctri.nic.in/Clinicaltrials/pmaindet2.php?trialid=12607. [Last cited 2020 Oct 02].
24.	Clinical Trials Registry India [Internet]. New Delhi: database publisher (India). 2007 Jun 20 - . Identifier CTRI/2019/02/017376, Assessing impact of vitamin D on kidney health; 2019. Available from: www.ctri.nic.in/Clinicaltrials/pmaindet2.php?trialid=31118. [Last cited on 2020 Oct 02].
25.	Alon U, Chan JC. Calcium and Vitamin D homeostasis in the nephrotic syndrome: Current status. Nephron 1984;36:1-4.
26.	Bacchetta J, Harambat J, Cochat P. Long-term steroid therapy in children: Is adjunct therapy relevant in nephrotic syndrome? Arch Pediatr 2008;15:1685-92.
27.	Strauss J, Zilleruelo G, Freundlich M, Abitol C. Less commonly recognized features of childhood nephrotic syndrome. Pediatr Clin North Am 1987;34:591-607.
28.	Singh DN, Krishnamurthy S, Kamalanathan SK, Harichandrakumar KT, Sivamurukan P. Three-monthly bolus vitamin D supplements (1000 vs 400 IU/day) for prevention of bone loss in children with difficult-to-treat nephrotic syndrome: A randomised clinical trial. Paediatr Int Child Health 2018;38:251-60.
29.	Chen Y, Wan JX, Jiang DW, Fu BB, Cui J, Li GF, et al. Efficacy of calcitriol in treating glucocorticoidinduced osteoporosis in patients with nephrotic syndrome: An open-label, randomized controlled study. Clin Nephrol 2015;84:262-9.
30.	Selewski DT, Chen A, Shatat IF, Pais P, Greenbaum LA, Geier P, et al. Vitamin D in incident nephrotic syndrome: A midwest pediatric nephrology consortium study. Pediatr Nephrol 2016;31:465-72.
31.	Basu S, Banerjee S, Sengupta J, Das S. A randomised controlled trial to assess the effect of vitamin D supplementation in steroid sensitive nephrotic syndrome. Indian J Clin Biochem 2015; 30 (Suppl 1):S1–S13.