Effect of dosage failed of first antiepileptic drug on subsequent outcome

The recent definition of drug-resistant epilepsy proposed by the International League Against Epilepsy (ILAE) stipulated failure of an adequate trial of two tolerated, appropriately chosen and used antiepileptic drug (AED) schedules to achieve seizure freedom. Doses failed were not specifically discussed. We explored the effect of the doses at which the first and second AED regimens failed on subsequent outcomes in a population of adults with newly diagnosed epilepsy followed for up to 20 years.

Patients in whom epilepsy was diagnosed and the first AED prescribed between July 1, 1982 and April 1, 2006, were followed until March 31, 2008. Dosage at which an AED failed was categorized according to the World Health Organization's defined daily dose (DDD) for each drug. Cumulative incidence curves for time to final seizure freedom (no seizure for at least 1 year on unchanged dosage at last follow up) were stratified by whether the first regimen was failed at doses above or below the 25%, 50%, or 75% cutoffs for the DDD of each AED.

Among patients who had taken a second regimen (n = 327), those in whom the first AED failed at doses above the various cutoffs (particularly 50% and 75% DDD) had lower probability of becoming seizure-free at last follow-up (p = 0.06 for 25% DDD, p < 0.001 for both 50% and 75% DDD). The same difference was observed for patients who had taken a third regimen (n = 141; p = 0.23 for 25% DDD, p < 0.01 for 50% DDD; and p = 0.002 for 75% DDD). A trend to higher seizure-free rate was observed in patients who had taken the third regimen when both the first and second regimens failed at <75% DDD. The difference remained significant after adjusting for covariates when using 50% DDD as the cutoff for patients who took a second regimen (hazard ratio 1.60, 95% confidence interval 1.08-2.37).

Higher failure dosage of the first AED predicts poorer subsequent outcome. This methodology could be used to refine further the ILAE definition of drug-resistant epilepsy by exploring the doses need to fail to provide an adequate AED trial.