We thank the authors of the letter 1 for their interest in our work 2 and for their insightful and interesting comments. Our study 2 is based on a national database, the Swedish National Epilepsy Surgery Register (SNESUR). This allows us to analyze data from several centers and to include a large number of cases and a wide range of epilepsy surgery procedures. This increases the generalizability of the results, and the prospective collection of predetermined variables reduces the risk of

more »

... ng bias. However, there are limitations inherent to this study design. Importantly, the analyses are limited to the variables defined in the register protocol. In the letter, 1 a number of ideas for further analyses are suggested: completeness of resection or disconnection, preoperative seizure semiology, change in pattern of seizures, and postoperative status epilepticus. We agree that such data would have been useful to further the understanding of possible reasons for worsening after epilepsy surgery, but unfortunately these variables are not included in SNESUR. Such detailed analyses might be more feasible for a retrospective study by a large epilepsy surgery center. When running a national register for a number of epilepsy surgery centers over a long period of time (30 yr and 6 centers for SNESUR), there is an inherent opposition between upholding the reporting quality of the register and the number of variables to enter for each patient. In order not to exhaust the reporting colleagues to a degree risking their participation, it is necessary to limit the number of variables to enter. Moreover, the integrity of long-term follow-up limits the possibility to incorporate new variables of interest as knowledge expands. In the manuscript, we propose that the opposite associations between preoperative seizure frequency and postoperative worsening reported by Sarkis et al 3 and in our study depend on differences in the surgical panorama. Still, we acknowledge that each proposed risk factor is explorative and has to be replicated in future studies. The suggestion to analyze long-term follow-up and its effect on seizure frequency is interesting and worthy of a separate study. Previous studies have shown that there are mixed trajectories of seizure freedom after epilepsy surgery, with relapses after initial seizure freedom as well as improvement over time in some patients who have recurrent seizures. 4,5 It would therefore certainly be informative to investigate if seizure worsening is stable over time.