Abstracts

Rationale: A potential association exists between epilepsy and impaired bone metabolism. It has been suggested that enzyme-inducing antiepileptic drugs (EIAEDs) affect CYP450 isoenzymes, potentially increasing vitamin D metabolism, resulting in reduced 25-OH vitamin D (25-OHD) plasma levels. It follows then that a potential pharmacokinetic interaction could exist between EIAEDs and oral formulations of vitamin D used for supplementation. Our objective was to characterize concomitant administration of AEDs upon vitamin D dose requirements. Methods: A retrospective chart review was conducted on all patients with epilepsy on any AED regimen, and were prescribed vitamin D from the Madison VA between 1/2012 and 9/2017. Patients were stratified into 2 groups based upon AED regimen: EIAED use or non-inducing AED (NIAED) use. Patients receiving polytherapy with both EIAED and NIAED were placed in the EIAED group. Patients were excluded from the analysis if they had a malabsorptive condition, or if they were using calcitriol. Data abstraction included: AED used, prescription and OTC vitamin D use, 25-OHD plasma concentration, renal function, age, gender, and ethnicity. 25-OHD was measured using a chemiluminescence immunoassay. A minimum 25-OHD plasma level of 30 ng/ml was considered to be the therapeutic goal. Within each patient, full data was abstracted for each 25-OHD level as a new observation throughout the evaluation period. All vitamin D doses were converted to a daily dose and assumed a 1 to 1 conversion for ergocalciferol to cholecalciferol units. Student’s t-test and Fischer’s Exact Test were initially used to compare patients using EIAED to those who were not. Linear regression was used to evaluate the difference in vitamin D dose by enzyme-inducing status while minimizing the effects of potentially confounding variables (25-OHD concentration, OTC vitamin D use, chronic kidney disease [CKD], age, gender, and ethnicity). Data was clustered by patient in the linear regression to account for multiple observations within the same patient. Results:  N=1,113 observations were derived from 315 patients. There were 263 observations in the EIAED group (23.6%). Of the EIAED group, 90.5% were male, vs 91.8% in NIAED. Mean age for EIAED vs NIAED patients was 65.9 & 61.4 yrs respectively (p=0.529). All variables were evenly distributed between EIAED and NIAED groups except: CKD (6.1% EIAED vs 13.8% NIAED; p<0.001), ethnicity (78.7% caucasian EIAED vs 87.7% NIAED;p<0.001). The most common EIAED was PHT (50.6%), followed by CBZ (31.9%), PB (14.1%), OXC (6.8%), PRM (1.9%), ESL (0.8%).  While the mean 25-OHD levels were similar between groups, and both groups met goal levels (32 ng/ml EIAED vs 33.2 ng/ml NIAED; p=0.28), a significant difference in the daily dose of vitamin D needed to achieve these levels was seen and differed by AED use (1587 units/d EIAED vs 1108 units/d NIAED;p-0.0002). Results of the linear regression were consistent with the primary analysis with a 409 unit increase in vitamin D dose among EIAED group (p=0.052). Conclusions: We found that patients taking EIAEDs, either alone or in combination, required a clinically meaningful increase in vitamin D dose in order to achieve the same 25-OHD plasma level as those taking NIAEDs. This suggests that patients taking EIAEDs may benefit from more intensive monitoring of their vitamin D supplementation, and clinicians should anticipate this likely pharmacokinetic interaction. Funding: None