INVITED REVIEWS Electrical Control of Epileptic Seizures Yue Li and David J. Mogul Summary: Epilepsy is among the most common neurologic disor- ders, yet it is estimated that about one third of patients do not respond favorably to currently available drug treatments and up to 50% experience major side effects of these treatments. Although surgical resection of seizure foci can provide reduction or cessation of seizure incidents, a significant fraction of pharmacologically intractable seizure patients are not considered viable candidates for such procedures. Research advances in applying electrical stimula- tion as an alternative treatment for intractable epilepsy have been reported. The primary focus of these studies has been the search for optimized stimulation protocols by which to electrically suppress, revert or prevent seizures. In this review, the authors discuss some of the promising results that have been achieved. These results are organized in three broad categories based on how such protocols are generated. They focus on how information of the electrical activity in the brain is incorporated in the control schemes, namely: open loop, semiclosed loop, and closed loop protocols. Benefits, potential promises, and challenges of these different control techniques are discussed. (J Clin Neurophysiol 2007;24: 197–204) E pilepsy is categorized as any of various disorders marked by repetitive aberrant electrical activity in the central nervous system and typically manifested by convulsive at- tacks known as seizures. It is estimated that approximately 1–2% of the world’s population may be afflicted with epi- lepsy, making it among the most common of neurologic disorders (Engel, 1989; Engel and Pedley, 1997). The hall- mark of epilepsy is recurrent seizures. Recurrent seizures may occur as a result of a large number of causes and the underlying mechanisms frequently are not fully understood. If seizures cannot be controlled, the patient may experience major disruptions in family, social, educational, and voca- tional activities that can have profound impacts on the quality of life (Goldstein and Harden, 2000). The mainstay of treatment is chronic medication based on modulation of cortical inhibition/excitation balance to prevent seizures. Anticonvulsant drugs help about two thirds of epilepsy patients achieve effective seizure control. The remaining approximately one third are refractory to pharma- ceutical therapy (Fisher, 1998). Furthermore, many patients develop a tolerance to the anticonvulsant effects, causing a marked decrease in drug efficacy. In addition, these drugs frequently have many concomitant side effects such as diz- ziness, drowsiness, impaired vision, headache, mood change, rash, and weight gain. For some patients, surgery may be another option by having the focus generating a partial seizure mapped and surgically removed. Although surgery is successful in pre- venting seizures in about 8% of the total epileptic patients (Engel et al., 1993; Tellez–Zenteno et al., 2005), there are legitimate fears of possible surgical complications and neu- rologic deficits such as memory loss or cognitive impairment. Also, the excision of an epileptogenic focus becomes imprac- tical or inefficacious in patients with multiple foci or with generalized seizures. Finally, surgery techniques are anatom- ically irreversible. Such intractable epilepsy, both resistant to drug treatment and unsuitable for surgery, is a significant public health problem so that other alternative therapeutic approaches are needed. For years, attempts have been made to apply electrical simulation for seizure control as an alternative treatment. Stimulation may have the advantages of reversibility and adjustability in patients who would otherwise be thought of as candidates for surgery. No brain tissue is destroyed and the stimulator can theoretically be adjusted to achieve the best outcome. It can also be turned off or removed if adverse side effects occur. In this review, we discuss some of the prom- ising results that have been achieved. These studies have been organized into three broad categories based on how these protocols are generated: open loop (predefined therapy, peri- odic stimulation independent of brain activity), semiclosed loop (predefined therapy, detected brain event as the trigger for stimulation), and closed loop (adaptive therapeutic proto- col, stimulation patterns and timing determined by an analy- sis of brain activity). OPEN LOOP STIMULATION “Open loop” refers to stimulation that is independent of brain activity at any particular point in time such that the dynamics of neuronal behavior are not incorporated in the generation of stimulation protocols. The therapy is based on a predefined schedule, independent of physiologic activity (Fig. 1). Such “open-loop” protocols typically stimulate in repetitive cycles of timed “on” and “off” periods. Pritzker Institute of Biomedical Science and Engineering, Illinois Institute of Technology, Chicago, Illinois, U.S.A. Address correspondence and reprint requests to David J. Mogul, PhD, Department of Biomedical Engineering, Illinois Institute of Technology, 10 West 32nd Street, Chicago, IL 60616, U.S.A.; e-mail: mogul@iit.edu. Copyright © 2007 by the American Clinical Neurophysiology Society ISSN: 0736-0258/07/2402-0197 Journal of Clinical Neurophysiology • Volume 24, Number 2, April 2007 197