Abstracts

Rationale: Knowledge regarding potential relationships between serum antiepileptic drug (AED) levels and tolerability can be helpful for epilepsy management. To date, there is limited data on the clinical utility of clobazam (CLB) serum levels. In order to better characterize the relationships between CLB dose, serum level, and toxicity, we examined outpatient records of 172 epilepsy patients treated with CLB.Methods: From an ongoing, retrospective, longitudinal observational study of AED response and tolerability, we identified CLB serum levels of patients treated with steady daily doses of CLB for seizure control. N-desmethylclobazam levels were not tracked. We omitted levels collected during pregnancies or medication titration periods. If a patient had multiple CLB levels collected during distinct time periods on the same daily dose and AED regimen, these levels were averaged. We then performed a correlational analysis to characterize the relationship between CLB serum level and daily dose, accounting for instances in which patients were concurrently taking enzyme-inducing AEDs (phenytoin, phenobarbital, and/or carbamazepine). To determine whether dose or level were related to adverse effects or toxicity (adverse effects prompting a dose decrease), we utilized linear discriminant analyses and a generalized estimating equation model given our longitudinal and correlated data. We stratified serum levels into four categories for the generalized estimating equation, using only one data point per patient at each level stratum.Results: 803 records of CLB serum levels from 172 patients (100 male) met our inclusion criteria. Mean age was 42.6 years (SD ± 14.5; Range = [16.3,78.6]) at the time of each patient’s first CLB record. We found a positive linear correlation between CLB dose and level (Pearson r = .226; p < .001) (Figure 1). CLB levels were lower while patients were taking concomitant enzyme-inducing AEDs (p < .001) despite mean doses that were 25.6% higher. Of 172 patients analyzed, 23 (13%) had adverse effects attributed to CLB, with 11 (6%) patients experiencing toxic adverse effects leading to a dose decrease. Drowsiness was the most common adverse effect, occurring in 16 patients. No other adverse effect was found in more than three patients. The linear discriminant analysis showed that dose was not a significant predictor of adverse effects or toxicity. The generalized estimating equation did not show a significant relationship between CLB serum level and the incidence of adverse effects or toxicity (Figure 2).Conclusions: We found that CLB serum level had a weak positive linear relationship with dose and was affected by the concomitant use of enzyme-inducing AEDs. We did not find any significant relationships between CLB serum level or dose versus the incidence of adverse effects or toxicity. While it is possible that no relationships exist, our results may have been due to our small sample, which was limited to patients with records of CLB serum levels. Other confounding variables, such as the concomitant use of non-AED enzyme inducers, may have also impacted our results.