Review Extracerebral detection of seizures: A new era in epileptology? I. Osorio a, ⁎, S. Schachter b a Department of Neurology, University of Kansas Medical Center, Kansas City, KS, USA b Comprehensive Epilepsy Center, Beth Israel Deaconess Medical Center, Boston, MA, USA abstract article info Article history: Accepted 9 September 2011 Keywords: Extracerebral signals Heart rate Electrocardiogram Motor Actimetry Accelerometers Seizure detection Real time The medical and psycho-socio-economic burden imposed on patients, caregivers, and health systems by pharmacoresistant epilepsies is enormous. Intracranial devices for automated detection, warning, and deliv- ery of therapy, the presently preferred “line of attack” for an abundance of weighty reasons, would be insuffi- cient to adequately address said burden on a global scale. Reliance on signals that, although extracerebral, are under cortical modulation or control and are altered by seizures, such as cardiac or motor signals, emerges as a viable research direction with potentially fruitful clinical applications. The greater ease of implementation and lower cost of automated real-time detection, warning, and therapy systems based on extracerebral signals, compared with those requiring intracranial placement, make them worthy of investigation. This article is part of a Supplemental Special Issue entitled The Future of Automated Seizure Detection and Prediction. © 2011 Elsevier Inc. All rights reserved. For self-evident reasons, cortical electrical activity has been the sole source of signals for visual or automated detection and quantifi- cation of seizures in clinical use. The inextricable link between brain and epilepsy has historically impelled clinical neuroscientists to leave unexploited the equally inextricable link between brain and body. The brain–epilepsy link has distracted us from certain severe limitations (for certain applications) inherent to the recording of cor- tical signals from scalp or even directly from its surface, such as marked cortical signal attenuation and filtering and limited access to neural sources (only about one-third of the neocortex is surveyable by scalp electrodes, and subdural electrodes record little activity from the lateral and bottom walls of sulci) [1,2]. Yet, readily accessible sources that provide indirect but valuable information about the state of the brain, particularly during the ictal or postictal state, remain largely untapped. The growing emphasis on widely accessible, cost-effective, good- quality health care in the context of expanding populations, especially in age groups above 60 years in whom the incidence of epilepsy is high [3], and the shrinking financial resources to support the required in- frastructure pose an enormous challenge to patients whose seizures are pharmacoresistant, as well as to epileptologists and functional neurosurgeons. The properly placed emphasis on implantable intra- cranial devices for automated seizure detection, warning, and delivery of therapy in patients with drug-resistant seizures should be viewed in the context that even if economic resources were unlimited, human re- sources are starkly small. Given the number of functional neurosur- geons in the United States (one source puts the number at 300, of whom about 100 work in epilepsy), is it realistic to pursue exclusively intracranial devices to address the unmet needs of pharmacoresistant patients conservatively estimated (in the United States) at 600,000 [4]? The deleterious medical and psychosocial impact of intractable epilepsy and its high cost of care [5], along with the sophisticated human and technological resources needed to address them, qualify this, in these authors’ opinion, as a public health care problem. Indeed, scientific advances, regardless of their value, may not translate into improved care of epilepsy and lessen its burden [6] unless devices are broadly accessible; in short, the challenge of ameliorating the global burden of drug-resistant epilepsies may exceed scientific and techno- logical ones. If the answer to the question put forth a few lines above is in the negative (intracranial devices will not meet the global burden), viable alternatives must be sought. The utilization of certain extracerebral signals looms as one such alternative. Cardiac (e.g., heart rate, EKG morphology) [7–19] and motor (speed, direction, and force of joint movements) [20] signals are prime candidates for the following reasons: (1) Structures that form part of the central autonomic nervous system [21–26] or are strongly interconnected with it are common sites of epileptogenesis (e.g., amygdalae, hippocampi). (2) Spread of seizures out of the prima- ry epileptogenic zone is prevalent in pharmacoresistant patients, so that even if the site of origin is not part of the central autonomic network, invasion of it by ictal activity is quite common [27]. (3) Partial seizures, particularly if complex, are characterized by either positive (e.g., motor automatisms, hypermotoric behavior, clonic/myoclonic Epilepsy & Behavior 22 (2011) S82–S87 ⁎ Corresponding author at: Department of Neurology, University of Kansas Medical Center, 390 Rainbow Boulevard, Kansas City, KS 66160, USA. Fax: +1 913 588 4585. E-mail address: iosorio@kumc.edu (I. Osorio). 1525-5050/$ – see front matter © 2011 Elsevier Inc. All rights reserved. doi:10.1016/j.yebeh.2011.09.012 Contents lists available at SciVerse ScienceDirect Epilepsy & Behavior journal homepage: www.elsevier.com/locate/yebeh