with phenobarbital, 3 with clonazepam, 2 with mesuximide, and 1 with ethadione.The results of biotin analysis in epileptics were compared with biotin plasma levels for 93 controls (blood donors and patients with dermatological diseases) who had a mean age of 42 years. Statistical analysis between groups was carried out using the Student t test.In 264 epileptics undergoing long-term therapy with anticonvulsants, significantly reduced plasma biotin levels were found compared with a normal control group: 74% of the epileptics had biotin levels of 250 ngIL or less. In patients undergoing single-drug therapy, the mean plasma biotin levels for those treated with sodium valproate were higher than for those treated with phenytoin, primidone, or carbamazepine. The observed reduction in biotin levels might be a factor influencing the efficacy of these three anticonvulsants.Krause K-H, Berlit P, Bonjour J-P: Impaired biotin status in anticonvulsant therapy. Ann Neurol 12: [485][486] 1982 Recently we reported reduced plasma biotin levels in epileptics undergoing long-term therapy with anticonvulsants [5]. The plasma biotin levels showed a significant negative correlation with the total amount and average daily dose of anticonvulsants administered. Because of the small number of patients examined, no differentiation of the effect of single anticonvulsants could be made. Subsequently, we have investigated 264 patients.
MethodsPlasma biotin concentrations were determined microbiologically using Lactobacillus plantarum as a test or- 0.001). The plasma biotin levels in patients taking these three anticonvulsants were significantly lower than in the controls (p < 0.0005 in each case).
DiscussionThe previously reported reduction of biotin plasma levels in epileptics undergoing anticonvulsant treatment [ 51 has been impressively affirmed. Considering that epileptics treated with phenytoin, primidone, o r carbamazepine had significantly lower plasma biotin levels than those treated with sodium valproate, the question arises whether the reduction in biotin levels might be a factor in the efficacy of these drugs. A reduction of the biotin-dependent pyruvate carboxylase may cause a reduced level of aspartate via the decrease of oxalacetate. A decline of this neurotransmitter in the brain, caused by phenytoin intake, has recently been demonstrated in animals [6]. Dermatitis and ataxia are clinical manifestations of an inborn error of biotin-dependent enzymes [l-3, 7, 81. It is of interest that these symptoms are wellknown side effects of anticonvulsants, with the exception of sodium valproate. On the other hand, seizures have been reported in some cases of an inborn error of biotin-dependent enzymes [l, 3 , 81. Altogether, the implication of the reported decline of biotin levels under anticonvulsant treatment is not clear. The pharmacological effect of biotin deficiency in epileptics deserves further attention.