Abstracts

Rationale: The relationship between carbamazepine use in patients with epilepsy and vitamin D deficiency has been studied, yet no comprehensive reviews to inform practitioners and policymakers is currently available.  Methods: We performed a meta-analysis on studies that measured hydroxyvitamin D (OHD) levels in persons taking carbamazepine compared to persons not on carbamazepine. We searched pubmed.org for the terms “carbamazepine” and “vitamin D” and found 103 titles, which we reviewed systematically to identify studies that measured OHD levels in patients on carbamazepine monotherapy compared to a control group. Of the 20 studies that fit these criteria, 12 had data reported in means and standard deviations with sample sizes for each group (Figure 1). The mean difference in OHD was standardized by calculating Cohen’s d. Effect sizes were calculated based on the difference between the two raw means divided by the pooled standard deviation. Results: The difference in means was -4.00 (95% CI = -7.12 to -0.88), meaning that on average, the CBZ groups had OHD levels 4.00 ng/mL lower than comparison groups (Figure 2). Meta-regression analysis revealed no significant effect on the observed OHD levels with regard to age (Z=0.26, p=0.79), sex (Z=0.72, p=0.47), or duration of CBZ therapy (Z=-1.04, p=0.30). Whether the studies explicitly excluded, included or did not explicitly address patients taking OHD supplementation did not significantly alter effect size (Q(2)=3.97, p=0.14). Studies that used a healthy control comparison group had a mean difference of -4.85 (95% CI = -8.15 to -1.55), and studies that used an epilepsy patient comparison group not taking CBZ had a mean difference of -0.23 (95% CI = -7.18 to 6.72). Mixed effect analysis indicated no statistically significant difference in effect sizes based on comparison group type (Q(1)=1.39, p=0.24). Conclusions: Carbamazepine leads to reduced OHD levels, independent of the duration of therapy. This meta-analysis suggests induction of Vitamin D metabolism by CBZ. Future analysis will compare vitamin D levels between enzyme-inducing AEDs and non-enzyme inducing AEDs. Funding: No funding