SIR-We thank Brunklaus et al. for the points they make regarding our study on the use of methylphenidate in patients with active or difficult-to-treat epilepsies. 1 These authors have produced recent data on the high prevalence of hyperactive and impulsive behaviours in patients with Dravet syndrome and severe epilepsy, and discuss the potential benefit of providing specific treatment, highlighting the relevance of the findings in our open-label trial. 2 The major deterrent for use of methylphenidate in epilepsy is the perceived risk of seizure worsening. But the main finding of our study was that this perception was not confirmed, even in patients in whom seizures were active. This was irrespective of whether several previous adjustments of antiepileptic drugs (AEDs) brought attacks under control. In our minds, this is certainly more relevant than the other finding of the study, i.e. that seizure control improved with methylphenidate. As Brunklaus et al. correctly suggest, the latter is more difficult to disentangle from the effects of AED adjustments and, we would add, the more intense care these patients received during the study; one of the caveats of open-label, non-controlled trials.In spite of the considerations above, we would like to mention that in only two patients (nos 2 and 11; Table SIII) were additional AED adjustments made (between <1mo before the start of methylphenidate and >1mo after the start of methylphenidate). Neither seizure frequency nor severity was affected by these adjustments. We also repeated the statistical analyses excluding these two patients and the results of the study were confirmed (data available upon request). Thus, in some patients, methylphenidate in the dose range we used apparently improved seizure control and this possibility should be kept in mind, particularly in light of recent data indicating that activation of dopamine D2 receptors may help control seizures in children. 3 Finally, we agree that we could have chosen a different term concerning the severity of the epilepsy. However, because even those patients who were on monotherapy or eventually achieved seizure control during AED adjustments had had a number of medication changes over the yearsand because we had the clinician in mind -we kept the term 'difficult-to-treat'. A number of specialists have been struggling with definitions of seizure refractoriness or drug-resistant epilepsy, 4 although agreement upon such definitions are important for a number of reasons. Clinicians can distinguish those patients who have seizures that are difficult to control from those in which they are unable to do so. Both scenarios have been similarly plagued by a longstanding perception that using stimulants may make things worse. We view our contribution as a preliminary demonstration that this may not be so, thus allowing consideration of specific treatment of disruptive behaviours that have a significant impact upon the quality of life of these children and their carers.