Kerala is a state located at the southern tip of India and receives adequate sunshine throughout the year. Consecutive ambulant children (aged 2-13 years) with epilepsy and having apparently normal physical and mental development, and attending the Pediatric Neurology outpatient department of a tertiary referral hospital in Kerala between June 2012 and May 2013, on either CBZ or VPA monotherapy for at least six months, were included in the study after taking informed consent. The study protocol was approved by the Institutional Research Board and ethical clearance was granted by the Institutional Ethical Committee. Children who received vitamin D or calcium supplementation, those on polytherapy with antiepileptic drugs (AEDs) or any chronic medications likely to affect bone metabolism like vitamin A, anabolic steroids, bisphosphonates, gluco-corticoids, thiazides, calcitonin etc. and children with history of malabsorption, hypothyroidism, hepatic or renal diseases were excluded from the study. Control group included age-matched children attending a nearby day-care centre and a school, who were not on any continuous medications during the same period of study.

25 (OH) D level was analyzed using ELISA 96T kit
(Diametra, Italy) within two weeks of sample collection. The lowest detectable concentration of 25 OH vitamin D is 0.3 ng/mL at 95% confidence limit. The intra assay variability was less than 6.4% and inter-assay variability was less than 6.95% (precision values as per the manufacturer). Children with serum level of 25 (OH) D level of more than 20 ng/mL (50 nmol/L) was considered sufficient, levels between 12-20 ng/mL (30-50 nmol/L) were considered insufficient, and below 12 ng/mL (<30 nmol/L) were considered deficient [4].

A priori power calculation based on a previous study [5] indicated that a sample size of 30 children in each group would provide 80% power to detect a 10% difference in 25(OH) D concentration, using a two-tailed t-test, while controlling type I error rate to 5%.

Statistical analysis: Comparison of quantitative data between two groups were analyzed by independent sample t test or Mann Whitney U test according to the nature of the data. Comparison of quantitative data among more than two groups were analysed by ANOVA with post hoc analysis or Kruskal Wallis test. Association between qualitative data were analyzed by Chi-square test. A P value of <0.05 was taken as statistically significant. Data analysis was performed using SPSS version 22.0.

56 children with epilepsy (28 each receiving CBZ and VPA monotherapy) and 109 controls were enrolled in the study. None of the cases or controls had fractures, bone pain, muscle pain or muscle weakness. Clinical features of rickets were absent in both cases and controls (Table I). Serum albumin and protein levels were in the normal range for all the three groups.

TABLE I Demographic Characteristics of Children Receiving Antiepileptic Drugs and Controls

Fig. 1 Box and whisker plot showing vitamin D levels in children on valproate Or carbamazepine monotherapy.

The median values of 25 (OH) vitamin D were 18.0 ng/mL (IQR 13.7-27.3), 21.3 ng/mL (IQR 16.4 -25.2) and 30.1 ng/mL (IQR 19.1-43.7) in the CBZ , VPA and control group, respectively (P=0.008). Comparison between CBZ and Controls (P=0.01) and VPA and Controls (P=0.02) also showed significant differences (Fig. 1). The proportion of participants with subnormal vitamin D levels (<20 ng/mL) were significantly different between the groups (Table II). Alkaline phosphatase levels were significantly higher among children on CBZ compared to controls (P=0.001) (Table III).

TABLE II Vitamin D Status of Cases and Controls 

TABLE III Calcium, Phosphorus and Alkaline Phosphatase Levels in Those Receiving Antiepileptic Drugs and Controls

This hospital-based cross-sectional study found a significantly high proportion of children receiving AEDs to have hypovitaminosis D, as compared to controls.

Fong, et al. [9] found that Indian ethnicity, immobility and polytherapy with AEDs were significant risk factors for low vitamin D levels in children with epilepsy. Seth, et al. [10] found that 83% of non-ambulant children with cerebral palsy on antiepileptic drugs were vitamin D deficient. Other authors have also reported low vitamin D levels in adults and children [11,12]. However, Turan, et al. [13] noted that CBZ, VPA and phenobarbitone therapy did not show any effect on serum vitamin D levels, as also reported with VPA in another study [14]. Hepatic induction of the cytochrome P450 enzyme system leading to increased catabolism of vitamin D is the principal mechanism reported in case of enzyme-inducing drugs like Carbamazepine [15]. Valproate inhibits the 25-hydroxylase activity on vitamin D in liver mitochondria without inhibiting the components of cytochrome P450-linked mono-oxygenase systems [16]. It is proposed that genetic variations like polymorphisms in vitamin D receptor (VDR) gene may predispose one to vitamin D deficiency [17]. The significant increase in serum alkaline phosphatase in children on carbamazepine may be attributed to changes in bone mineral metabolism due to its enzyme inducing property [18]. Mikati, et al. [19] studied the effect of low dose vs high dose vitamin D in ambulatory adults and children on antiepileptic drugs and found that high-dose vitamin D therapy substantially increased bone mineral density at several skeletal sites in adults. In children, both doses resulted in comparable increases in bone mass.

This study shows that serum 25 OH vitamin D levels are significantly low in children on carbamazepine or valproate monotherapy. Children on antiepileptic drugs should have regular monitoring of Vitamin D levels, and/or supplementation with calcium and vitamin D even in children with normal growth and development, no limitation of physical activity and adequate exposure to sunshine. The impact of antiepileptic drugs on bone health is to be addressed by all Pediatricians, as early identification of vitamin D deficiency and supplementation of calcium and vitamin D can help majority of children on long term anticonvulsants.

Contributors: MS,KD: conceptualised the idea, design, collection of data, analysis, preparation of manuscript, review of literature; PAMK: supervised the study and edited the manuscript; BA: collection and analysis of data of the control population; JP: statistical analysis and preparation of tables and figures; MV: biochemical analysis and interpretation of data; SB: supervised the biochemical analysis.

Funding: A research grant from SAT Hospital endowment fund. Competing interest: None stated.