We undertook a systematic review of the world literature to assess the level of RCT activity in epilepsy surgery. Previous personal experience with RCTs in epilepsy surgery and examples of successful Canadian multicentre research networks

methods to deal with bias and confounders in clinical research. Among the many types of biases, two can dramatically influence the study results and are particularly important in assessing the effect of surgical interventions. Selection bias refers to the process by which participants are not selected equally for the treatment and the control groups, resulting in groups with different prognostic variables. The best way to avoid selection bias is through concealed randomization, a process by which each study participant has an equal chance of being allocated to the experimental or to the control group. The second type of bias ABSTRACT: Background: Although randomized controlled trials (RCTs) are the gold standard for evaluating therapeutic interventions, surgical RCTs are particularly challenging and few have been done in the field of epilepsy surgery. We assess the level of RCT activity in epilepsy surgery and propose feasible alternatives to develop sustainable research initiatives in this area. Methods: We undertook a systematic review of the world literature to assess the level of RCT activity in epilepsy surgery. Previous personal experience with RCTs in epilepsy surgery and examples of successful Canadian multicentre research networks were reviewed to propose initiatives for sustainable, valid research in epilepsy surgery. Results: We identified 12 RCTs in epilepsy surgery, including 692 patients, of whom 416 were involved in vagus nerve stimulation, 16 in various brain electrostimulation procedures, 180 in comparisons of different surgical techniques, and 80 in a comparison of medical versus surgical therapy. Most studies were of short duration (median = 3 months, range 3-12 months). In the area of resective surgery, only temporal lobe epilepsy has been subjected to any type of RCT comparison. All RCTs have been done within the last 13 years. There were no multicentre Canadian surgical studies. Conclusions: The adoption of RCTs in epilepsy surgery has been slow and difficult worldwide. Because of its universal health care system and its well established epilepsy surgery centres, Canada is in a strong position to create a national epilepsy surgery research initiative capable of undertaking high quality, sustainable research in epilepsy surgery.