Quantifying the response to antiepileptic drugs: effect of past treatment history.

OBJECTIVE: To quantify the response to treatment with antiepileptic drugs (AEDs) as a function of the past treatment history and identify additional prognostic factors for predicting the response to newly administered AED treatments.
METHODS: A cohort of 478 consecutive patients who received newly administered AED treatments between January 1999 and December 2004 and were followed prospectively for 1.5 to 7.5 years in a single epilepsy clinic.
RESULTS: The response to newly administered AED treatments was highly dependent on the past treatment history. The seizure-free rates decreased from 61.8% for the first AED to 41.7%, 16.6%, and 0% after one, two to five, and six to seven past AEDs proved inefficient. This response curve corresponded to a mono-exponential function with a maximal response of 61.8% and half-decay constant of 1.5 AEDs. Likewise the response curve describing a greater than 50% reduction in seizure frequency corresponded to a mono-exponential function with a maximal response of 85.3% and half-decay constant of two AEDs. Three additional independent prognostic factors for predicting the response to AEDs were identified: type of epilepsy, duration of epilepsy, and number of seizures in the 3 months prior to AED initiation.
CONCLUSION: Drug resistance is a graded process that follows a mono-exponential course with a half-decay constant of 1.5 to two antiepileptic drugs (AEDs). Although relative drug-resistant epilepsy can be diagnosed after failure of two past AEDs, absolute drug resistance requires failure of six AEDs, as a significant minority of patients (16.6%) is rendered seizure-free by addition of newly administered AEDs even after failure of two to five past antiepileptic drugs.