Computerized tremor analysis of valproate-induced tremor: a comparative study of controlled-release versus...

Valproate (VPA) induces postural tremor in 6-45% of patients. The characteristics of VPA-induced tremor have not yet been quantitatively assessed, and it is not known whether tremor prevalence or severity is affected by VPA formulation (controlled-release CR-VPA vs. conventional VPA). The aim of this study was quantitatively to assess tremor in epilepsy patients receiving VPA and to compare the effects of two VPA formulations (CR-VPA vs. VPA) on tremor severity.

In a prospective study, 18 consecutive patients with newly diagnosed focal or generalized epilepsy were assigned to receive alternately either VPA (n=10) or CR-VPA (n=8) monotherapy. Computerized tremor analysis was performed at baseline 1 day before initiating VPA treatment and repeated after a seizure-free period of >or=8 weeks, during which VPA doses had remained stable. Rest and postural tremor were recorded by accelerometry, and surface electromyograms (EMGs) were recorded from the wrist flexors and extensors.

At baseline, the two groups had similar postural tremor amplitudes. At follow-up, the CR-VPA group had remained at the same level, whereas VPA subjects exhibited a significant increase in tremor amplitudes (p<0.05) despite comparable VPA doses and comparable plasma VPA concentrations at the time of tremor testing.

This is the first study to assess quantitatively VPA-induced tremor by standardized tremor analysis. These results suggest that CR-VPA may cause less tremorigenic activity as compared with standard VPA. The mechanisms underlying this difference are unclear but may include greater peak-trough variation with VPA than with CR-VPA.