Risk factors for valproic acid resistance in childhood absence epilepsy

AIMS Valproic acid (VPA) is reported to be effective for the control of absence seizures in 75% of children. The aim of this study was to determine the clinical and socio-demographic factors associated with VPA response in newly diagnosed childhood absence epilepsy (CAE) and to determine if these factors also influence the chances of achieving long-term seizure freedom. METHODS Medical charts of 180 children with CAE were retrospectively reviewed. Clinical, electroencephalographic and imaging findings were recorded to correlate with complete VPA response and long-term epilepsy outcome. Factors associated with non-responsiveness were identified individually and in a multivariable logistic regression analysis. RESULTS Treatment was successful in 112 (58.3%) children. More children that were non-responsive to VPA experienced generalized tonic clonic seizures (GTCS) (33.8% vs. 13.4% for responders; p=0.001) and 52.9% had a pre-treatment seizure frequency greater than 10/day (vs. 27.0% for responders; p<0.001). Finally, responders were older at time of diagnosis versus non-responders (p=0.001). Absence of long-term seizure freedom was linked to the presence of GTCS, the absence of initial response and the need for multiple AEDs to control seizures. INTERPRETATION Our results suggest that clinical phenotypes are associated with reduced response rates to VPA. This should be taken into account when counselling families of children with newly diagnosed absence epilepsy.