The mean 25 hydroxy vitamin D levels in our study at enrolment was 11.27(6.08) ng/ml which falls under the definition of deficiency of vitamin D according to IAP guidelines [8], and the mean vitamin D during remission was 13.65(6.24) suggestive of vitamin D insufficiency. In a study by Cetin N et al the mean Vitamin D level during remission was 16.4 ( 9.09) in children with infrequently relapsing nephrotic syndrome after a mean steroid intake of 1 year [9]. Lower levels of ionized calcium, high PTH and high ALP in a setting of vitamin D deficiency suggests biochemical abnormalities related to abnormal bone mineral metabolism and ill effects of bone health. They demonstrated that even 12 weeks of glucocorticoid therapy acts as a risk factor for osteoporosis. The risk of osteoporosis was not dependent on cumulative steroid dose. The management of FENS consist of 12 weeks of glucocorticoid therapy, hence it is ideal to get vitamin D testing at baseline and supplement children with deficiency to prevent its ill effects on bone health. Choudhary et al, emphasized on routine calcium and vitamin D supplementation irrespective of vitamin D status for osteoprotection in children with nephrotic syndrome[10] .We found that healthy controls had insufficient 25 OH vitamin D (15.97 ± 7.01) ng/ml. Children with FENS failed to achieve levels equal to healthy controls and remained 25 OH vitamin D deficient ( 13.65 ± 6.24). India despite being a tropical country, with good sunlight exposure, our healthy children were also vitamin D insufficient.
Surve et al, assessed the influence of VDBP levels on vitamin D status of under-five healthy children and showed that physiologic variations in PTH and VDBP are to be considered in children before deciding treatment strategies. PTH and VDBP were considered as non-modifiable risk factors for vitamin D deficiency in children especially those under the age of 5 years. The above study also emphasised on the free hormone hypothesis and importance of measuring free and bioavailable vitamin D, rather than total 25 hydroxy vitamin D, to prevent underestimation and overestimation of vitamin D status in children[11]. The mean serum VDBP level in our children with FENS during relapse, was 242.90 ( 127.75) µg/ml. This was comparable to the mean value of 210.0 (137.40 ) µg/ml in a study by Aggarwal et al, in adults with idiopathic nephrotic syndrome [3]. In children explanation for significantly lower levels compared to controls could be higher levels of proteinuria, beyond the compensatory capacity of liver and immaturity of liver’s ability to produce VDBP at higher concentrations On the follow up visit at 4 weeks of remission the VDBP levels increased to a mean value of 550.70 (219.7)µg/ml, which was fairly comparable with mean of 616 (255.03)µg/ml seen in controls. The plausible explanation for preserved VDBP as with few previous studies could be, compensatory increase in VDBP synthesis by liver and genetic variation in gene coding for VDBP. Assessment of VDBP in nephrotic syndrome gets paramount importance in children with SRNS, FRNS and SDNS, and continue to have proteinuria or remain in partial remission for prolonged periods. There was not much difference in the levels of VDBP in male and female gender. Waldron et al, showed that VDBP acts a negative acute phase reactant, but we did not find any difference in VDBP levels in children with or without infection at the onset of disease [12]. In our study serum VDBP and serum albumin showed statistically significant positive correlation (r = 0.37, p < 0.05). For every one unit decrease in albumin, VDBP decreases by 117.9 units. This is similar to the study by Aggarwal et al, showing similar correlation ( r = 0.264, p = 0.008) [3]. Hence, in comparison with healthy children, children with FENS, were hypoalbuminemic, and had reduced VDBP, therefore its assessment for knowing exact bioavailable vitamin D status gains importance. Our study also gives the normative values of bioavailable vitamin D levels in healthy Indian children were data is lacking.
According to free hormone hypothesis, it is the free form of the hormone which is active rather than the total 25 hydroxy vitamin D. Therefore we calculated the free hydroxy vitamin D from serum albumin, VDBP and 25 hydroxy vitamin D. As the serum albumin level increased during remission, the free form of vitamin D decreased. Hence it would be ideal to measure free vitamin D levels during FENS and relapse to know the exact vitamin D status in children with FENS. The fraction of 25 OH vitamin D which is loosely bound to albumin is also available for action at target sites, therefore bioavailable vitamin D levels should be measured in children with nephrotic syndrome. It was observed in our study that the mean bioavailable vitamin D increases with increase in age. This was in conjunction to decrease in VDBP with increase in age. The bioavailable vitamin D levels of healthy controls (0.95 ± 0.25 ng/ml), were comparable to children with FENS at 4 weeks of remission (1.11 ± 0.84 ng/ml). This shows that bioavailable vitamin D is especially useful to study vitamin D status in children during relapse. In our study bioavailable vitamin D showed a significant correlation with serum albumin, VDBP and 25 OH vitamin D as with the study by Aggarwal et al [3]. However bioavailable vitamin D did not show better correlation with PTH than total 25 hydroxy vitamin D.
In a study by Denburg et al, in pediatric CKD secondary to glomerular disorders having nephrotic range proteinuria had lower levels of free and bioavailable vitamin D compared to other aetiology of CKD, supporting the evidence that bioavailable vitamin D is a very good marker of vitamin D status in children with proteinuria [4]. They found mean bioavailable vitamin D in children with CKD secondary to FSGS, with significant proteinuria to be 0.8 ng/ml and those secondary to CAKUT was 3.4 ng/ml. This was in correlation with mean bioavailable vitamin D level of 0.75 ng/ml in children with FENS, as with our study.
The important limitation of first known study in our country by Aggarwal et al, was that followup at remission was not done. In our study we followed patients in remission which showed near normalization of bioavailable vitamin D levels like healthy children [3]. In the study by Aggarwal et al, it was found that bone mineral density in patients with nephrotic syndrome correlated well with bioavailable vitamin D ( r = 0.358, p = 0.0002) than total 25 hydroxy vitamin D ( r = 0.174, p value = 0.079 ), We did not perform DEXA scan in children with FENS due to logistic issues which was a limitation. Other limitations were smaller sample size and shorter period of follow-up.