Abstract
Juvenile myoclonic epilepsy (JME) has been considered to be a discrete phenotype. Evidence from segregation analysis and twin studies suggest that JME is a genetically determined disease, and two previous linkage studies of separately ascertained kindreds have suggested that a JME locus, EJM 1, resides on chromosome 6p close to the HLA region. Regarding the treatment of JME, it is commonly accepted that sodium valproate (VPA) is the drug of first choice. However, there are no previous controlled studies of VPA dose- or concentration-effect relationship regarding JME.
We have performed linkage studies of 25 nuclear families with a JME proband and JME or idiopathic generalised epilepsy in at least one additional pedigree member with markers for the proposed EJM 1 region. In our set of JME families, there was no evidence for linkage at the HLA on 6p, and exclusion data were obtained for a region 10-30 cM telomeric ta the HLA the extent of which varied with the penetrance assumed. The contrast between our and the previous findings indicate that genetic heterogeneity exists within the JME syndrome. In a second linkage study with the same family set, we could exclude the possibility that JME and the Unverricht- Lundborg disease are allelic variants.
In a randomised, double-blind, cross-over dose-effect study using 500 mg and 1000 mg VPA b.i.d., we found a coefficient of variation for intraindividual repeated blood levels of about 20 percent. Hence, the pharmacokinetic requirements for VPA therapeutic drug monitoring are fulfilled. However, no correlation was found between VPA dose or concentration and seizure control in our 16 patients, and less than half of the patients were seizure-free despite VPA concentrations within the suggested therapeutic range. Indeed, 4 patients had more seizures during high dose treatment, but the sample is too small to reach statistical significance.
We found no clear concentration-dependent side-effects, and a series of neurophysiological tests did not change between low and high dose treatment. If these observation can be confirmed in further studies, the common clinical strategy to increase the dose until the patient is seizure-free or suffer side-effects, should be reconsidered. A 24 hour EEG was free from epileptiform activity in 75 percent during low dose and 69 percent during high dose treatment, also in patients not seizure-free. No change in background EEG activity was found between the two VPA doses. This indicates that EEG must be interpreted with care when used for phenotype classification and is of limited value for monitoring treatment effects of VPA.
