To the Editors: Recently, a Task Force appointed by the International League Against Epilepsy (ILAE) formulated a consensus definition of drug-resistant epilepsy as failure of two antiepileptic drug (AED) trials (Kwan et al., 2010). The Task Force also set standards to assess the response to AED treatments. Therefore, AEDs had to be appropriately selected, adequately used, and their failure due to lack of efficacy. A classic study conducted by Kwan and Brodie (2000) gave support to the consensus definition. In this study, 47% of the patients became seizure free after the first AED trial, whereas 13% attained this outcome after the second one. When the first two drugs had failed, only 4% achieved seizure freedom. However, the standards established by the Task Force could challenge these results. In Kwan and Brodie’s study, the discontinuation of AEDs caused by side effects or by reasons other than inefficacy was considered as drug failure. If we exclude from the analysis the 135 patients with a lack of response due to the aforementioned reasons, the outcome after the first informative AED trial would be 222 patients (66.2%) seizure free and 113 (33.7%) not seizure free. These rates are similar to those of another recent prospective cohort study (Schiller & Najjar, 2008). In this second study, when drug discontinuation due to adverse effects was included in the analysis, only 47.9% of patients became seizure-free with the first-ever tried AED. However, when only lack of efficacy was required for treatment failure, then, 61.8% achieved seizure freedom with their first AED. Using this approach, the rates of patients rendered seizure free with successive AED treatments were 41.7% for the second, 16.6% for the third to sixth, and 0% for further AED trials. These results could challenge the limit of two AED treatment failures included in the ILAE consensus definition. In fact, Schiller and Najjar found that the response curve to sequential AED trials corresponded to a monoexponential function with a maximal response of 61.8% and a half-decay constant of 1.5 AEDs. Drug resistance seems to be a graded process (Perucca, 1998; Schiller & Najjar, 2008). Therefore, the application of a graded system to categorize drug resistance could be appropriate (Perucca, 1998). We propose three grades of drug resistance depending on the number of previously failed AED trials: grade I, after two failures, grade II, after three to five failures, and grade III, after six or more failures. The Task Force recommended the use of their definition in scenarios that entail different degree of risk, such as the referral of patients from a general physician to an epilepsy specialist or the consideration of epilepsy surgery. These distinct situations could be better addressed through a graded approach. Just as illustrative examples, grade I drug resistance could be enough to refer a patient to an epilepsy specialist, grade II might be a requirement for a patient to participate in experimental AED trials, and, finally, grade III might be desirable for a patient to undertake a presurgical evaluation. The use of this graded system in clinical research would offer more information about the degree of refractoriness in patients than using the single limit established by the consensus definition. Dr. Gomez-Alonso has served on the advisory boards or received speaking honoraria or research grants from Eisai, GlaxoSmithKline, Janssen-Cilag, Novartis, Pfizer, Sanofi-Aventis, and UCB Pharma. Dr. Gil-Nagel has served in the advisory board or received speaker honoraria or research grants from Bial, Cyberonics, Eisai, GlaxoSmithKline, Janssen-Cilag, Medtronic, UCB-Pharma, and Valeant. We confirm that we have read the Journal’s position on issues involved in ethical publication and affirm that this report is consistent with those guidelines.