Introduction Posttraumatic epilepsy: Treatable epileptogenesis

Traumatic brain injury (TBI) is a major cause of acquired epilepsy, and can exacerbate seizure severity in individuals with preexisting epilepsy. TBI is one of the clearest examples of the process of epileptogenesis in neurology. The present supplement reflects the Merritt-Putnam Symposium offered at the 2007 annual meeting of the American Epilepsy Society, in Philadelphia, Pennsylvania. The charge of this symposium was to evaluate current evidence for epileptogenesis following TBI, to discuss new diagnostic methodology involving neuroimaging and neuroradiologic techniques, and to evaluate the status of treatment for this disorder, merging new data from experimental models with those from clinical studies. The presentations highlighted this disorder as a likely treatable cause of epilepsy, and an opportunity for a better understanding of epileptogenesis in general.

TBI is occurring with increasing frequency in today’s battle theater. Although war-related mortality is declining because of more sophisticated armored protection, head injury is present in an increasing number of survivors. The resulting neurologic impairments from open and closed head injury range in severity: from severe paralysis and major mental impairment to a high incidence of up to 50% of more subtle cognitive impairment such as posttraumatic stress disorder (PTSD). Epilepsy is another neurologic consequence of TBI, and overt seizures are reported in up to 50% of survivors (Lowenstein, 2009). Importantly, posttraumatic epilepsy (PTE) is a major factor in the inability of survivors of head injury to return to their pre-existing lifestyles and employment. To date, clinical trials aimed at prevention of epilepsy following TBI have failed (Temkin, 2009).

Clinical studies reveal that TBI is one of only a few undisputed examples of epileptogenesis in the human brain. Epileptogenesis refers to the process whereby nonepileptic brain is transformed into one that generates unprovoked seizures. In addition, epileptogenesis refers to the development and extension of brain tissue capable of generating chronic, recurrent, spontaneous behavioral and/or electrographic seizures. The process may start with an initial insult that may or may not involve acute seizure activity, but that lead to later development of epilepsy. Both experimental animal models and human observations have revealed that there is often a “latent” period following the initial insult during which there are no acute seizures, prior to the eventual emergence of spontaneous seizures (epilepsy) (Fig. 1A). In the case of TBI, the latency can be up to several years (Lowenstein, 2009).

Figure 1.

Time course of epileptogenesis. (A) An initial insult, such as traumatic brain injury (TBI) and/or status epilepticus occurs, followed by a “latent period” lasting weeks to months or even years prior to the onset of spontaneous seizures. This “latent period” represents a period during which a cascade of molecular and cellular events alters network excitability to result in spontaneous epileptiform activity. This “latent period” is also an opportunity for biomarker development and therapeutic intervention. (B) The cascade of events that are presently suggested by experimental evidence can be classified temporally following the initial insult. Early changes occur within seconds to minutes including induction of immediate early genes and posttranslational modification of receptor and ion-channel related proteins. Within hours to days, there can be neuronal death, inflammation, and altered transcriptional regulation of genes such as growth factors. A later phase lasting weeks to months includes morphologic alterations such as mossy fiber sprouting, gliosis, and neurogenesis.

Epilepsia © ILAE

The existence of a latent period prior to onset of epilepsy raises multiple important issues for diagnosis and treatment in the TBI population. Identification of the cellular and molecular changes involved in the cascade of events leading up to epilepsy might reveal new therapeutic targets (Fig. 1B). Multiple experimental models are revealing that there may be stepwise changes that occur in neuronal network over days to weeks or even months and years following an epileptogenic insult (Fig. 1B). Early changes include the induction of immediate early genes and posttranslational modifications of neurotransmitter receptor and ion channel/transporter proteins (McNamara et al., 2006; Cornejo et al., 2007; Rakhade et al., 2008). Within days, neuronal death, initiation of an inflammatory cascade, and new gene transcription has been reported to occur (Vezzani & Granata, 2005; Scharfman, 2007). Later changes occurring over days to weeks include anatomic changes including axonal sprouting and dendritic modifications, such as mossy fiber sprouting that is commonly observed as a hallmark of chronic epileptic brain (Dudek & Sutula, 2007). Hence, recent basic research suggests that there may be multiple intervention points for therapeutic prevention of epilepsy. Despite these encouraging observations, there are no pharmacologic or nonpharmacologic therapies available today that are truly antiepileptogenic.

Clinical trials show that treatment with conventional antiepileptic drugs (AEDs) following TBI does not protect against later development of epilepsy (Temkin, 2009). Hence, new strategies need to be developed that are aimed at specific factors within the epileptic cascade. PTE is one of the most amenable human epilepsy syndromes for application of new therapies that are developed in experimental models. There is a discrete inciting event in an otherwise normal brain, an opportunity to monitor and screen at desired intervals following the injury. Until recently, the PTE endophenotype was poorly understood. Modern imaging and neurophysiologic techniques are being applied to TBI to document course and progression (Diaz-Arrastia et al., 2009). An ultimate aim is to develop biomarkers that can predict the risk and onset of epilepsy in this important population. New advances in neurophysiologic assessment are revealing preepileptic signatures to damaged brain (Diaz-Arrastia, et al., 2009; Dichter, 2009). New imaging modalities including functional magnetic resonance imaging (MRI) and diffusion tensor imaging (DTI) are showing early changes as they progress and correlate with the emergence of epileptogenesis following TBI (Diaz-Arrastia et al., 2009).

Simultaneous with the development of new clinical diagnostic methods, animal models are being developed that emulate the human condition. Similar to the human, TBI in rodents appears to induce epilepsy at much delayed time points. In adult rats and mice, brain trauma produces discernable lesions that can be followed by MRI, progressive neuronal hyperexcitability, and seizure susceptibility. Many of the models fail to show spontaneous seizures, although recent work has generated lateral fluid percussion injury that results in later spontaneous seizures (Pitkanen et al., 2009). It is important to note that these rodent models also result in a cognitive impairment, which is a common comorbid condition with PTE. Hence these rodent models are suitable for preclinical testing of both diagnostic and predictive technologies as well as interventional therapies. Treatment trials with varying degrees of success have included compounds that block glutamate receptors or calcium channels, caspase inhibitors and antiapoptotic agents, and neurotrophic factors as well as stem cell transplantation (Pitkanen et al., 2009). Although there have been some positive results, these treatments have not yet been translated to the human. An important reason for this disconnect is (1) the lack of available biomarkers in clinical use to indicate underlying pathogenesis, and (2) the unclear alignment between animal models and human patients with respect to the time-dependent mechanisms within the epileptogenic cascade.

Perhaps the most direct approach to measure changes taking place in neuronal networks following injury is electrophysiologic recording. Here, precise alterations in synaptic circuitry can be assessed, and observations may more precisely delineate functional changes that precede the onset of epilepsy. A number of animal models have been examined with a variety of in vivo depth electrode recordings or ex vivo brain slice extra-cellular, intracellular, and whole cell patch clamp recordings (Prince et al., 2009). A common observation is impairment in γ-aminobutyric acid (GABA)ergic inhibition, and an impairment or loss of GABAergic interneurons (Prince et al., this supplement). Other investigators have shown alterations in glutamate and excitatory glutamatergic synapses that drive epileptiform events (Willmore & Ueda, 2008; Rakhade et al., 2008; Cornejo et al., 2007). An advantage of ex vivo slices is that cells can be filled following recordings and evaluated for protein and gene expression for structure–function correlations. Cell-specific therapies can also be applied in vivo or in vitro for specific proof-of-principle preclinical study (Prince et al., 2009).

Taken together, the emerging preclinical and clinical research bases are poised for larger scale clinical trial in PTE prevention within the next decade. PTE may be the most tractable forms of epileptogenesis. Currently, there are only two trials addressing antiepileptogenesis using currently available agents (Dichter, 2009). There is a need for better EEG screening of at-risk populations following TBI, and more certain electrophysiologic biomarkers for epilepsy risk. Imaging techniques also must be developed to identify patients at greatest risk for PTE. Basic research is generating new data concerning accessible intervention points within the epileptogenic cascade following TBI.

Of the broad spectrum of progressive epilepsies, PTE is a syndrome that clearly has one of the highest likelihoods of being a case where a “cure” can be developed. The following supplement addresses the state of our knowledge regarding the clinical epidemiology regarding PTE, past experience with therapeutic trials, novel neurophysiologic and neuroimaging technologies that are being applied in the clinic, basic research that is revealing novel mechanisms within the epileptogenic cascade and preclinical therapeutic trials, and finally, the future directions available for clinical trial development in this important population, both military and civilian.