Epileptic encephalopathies (EEs) refer to a group of severe epileptic disorders characterized by spontaneous, recurrent seizures that result in cognitive and behavioral disturbances (Avanzini et al., 2012; McTaque & Cross, 2013). In most cases, EEs represent challenging, difficult-to-treat conditions associated with unfavorable outcome (Dulac, 2001a). Despite the significance, little is known about the pathogenetic mechanisms, and in particular, about neuronal networks underlying EE. One important feature of most EEs is that different etiologic factors (genetic background, lesions, malformations, neurometabolic abnormalities, and so on) result in a similar clinical expression (similar type of seizures, similar interictal abnormalities). It may be suggested that different etiologies, depending on the developmental stage, engage encephalopathy-specific pathogenetic systems responsible for each particular EE. Recently, this hypothesis has been further developed by some authors that proposed the term of “system epilepsies” (Capovilla et al., 2009; Avanzini et al., 2012). EEs, in general, have the ideal profile of system epilepsies and can be considered the result of the enduring propensity to generate seizures in different brain areas that, alone, are unable to create a complex epileptic syndrome. Based on this suggestion, different studies on functional neuroimaging have tried to specify fingerprints of EEs on the network level, which would explain the clinical similarity despite the etiologic heterogeneity (for review see Laufs, 2012; Moeller et al., 2013). Moreover, little is known about mechanisms that describe how epileptic activity interferes with cognitive networks and contribute to the development of EE. Neurophysiologic and functional neuroimaging evidence suggests that epileptiform discharges may affect cognition through either transient effects on information processing in the brain, or through more long-lasting effects leading to prolonged inhibition of brain areas distant from but connected with the epileptic focus (Van Bogaert et al., 2012; McTaque & Cross, 2013). The evidence from neuroimaging studies are summarized in the present review.
Epileptic encephalopathies (EEs) represent a group of severe epileptic disorders associated with cognitive and behavioral disturbances. The mechanisms of cognitive disability in EEs remain unclear. This review summarized neuroimaging studies that have tried to describe specific fingerprints of brain activation in EE. Although the epileptic activity can be generated individually in different brain regions, it seems likely that the activity propagates in a syndrome-specific way. In some EEs, the epileptiform discharges were associated with an interruption of activity in the default mode network. In another EE, other mechanisms seem to underlie cognitive disability associated with epileptic activity, for example, abnormal connectivity pattern or interfering activity in the thalamocortical network. Further neuroimaging studies are needed to investigate the short-term and long-term impact of epileptic activity on cognition and development.
Neuroimaging of Cognitive Deficits in Epileptic Encephalopathies
One of the possible mechanisms that may explain cognitive deficits associated with epileptiform activity is related to the interaction between the epileptic and cognitive networks. In that context, the interference of epileptic activity with the default mode network (DMN) has been investigated in detail. The DMN consisting of precuneus, retrosplenial cortex, and parietal and anterior medial frontal cortex, and is active in the resting brain with a high degree of functional connectivity (Raichle et al., 2001). It has been suggested that the DMN constitutes a necessary favorable neurometabolic environment for cognitive functions, represents a physiologic baseline for processes of attention and working memory, and supports dynamic integration of cognitive and emotional processing (Raichle & Mintun, 2006). Abnormal activity in the DMN and disturbed connectivity between the structures involved may influence task performance and contribute to pathogenesis of neuropsychiatric disorders such as attention-deficit/hyperactivity disorder, Alzheimer's disease, autism, schizophrenia, and depression (Eichele et al., 2008; Broyd et al., 2009). Moreover, altered activity in the DMN has been associated with fluctuations and disturbance of consciousness (Boly et al., 2008). It has been suggested that disruption of the resting state activity by pathologic processes (e.g., those that give rise to epileptic spike) may be related to alterations in cognitive function and that this may be a possible mechanism that underlies cognitive deficits in epilepsy (Gotman et al., 2005). Deactivations in the DMN have been described in awake patients with primary and secondary generalized paroxysms and absence seizures (Aghakhani et al., 2004; Gotman et al., 2005; Hamandi et al., 2006; Moeller et al., 2008a,b, 2010a,b). These DMN deactivations may reflect disturbance of awareness or consciousness (Gotman et al., 2005). The abnormal connectivity in the DMN has been observed in several epileptic disorders (Zhang et al., 2009, 2010), including typical EE such as continuous spikes and waves during slow sleep (CSWS) and Lennox-Gastaut syndrome (Siniatchkin et al., 2010; Pizoli et al., 2011; Moehring et al., 2013; Moehring J, Kroeher B, Galka A, von Ondarza G, Moeller F, Wolff S, Granert O, Jansen O, Boor R, Stephani U, Siniatchkin M, unpublished manuscript). However, the abnormal activity and connectivity in the DMN is only one explanation among others for mechanisms of EE. The disturbed connectivity in other cognitive networks in relation to epileptic activity has been described in several studies (Bettus et al., 2009). However, there is an urgent need for further studies on functional neuroimaging that should describe short-term and long-term impact of epileptic activity on cognitive networks, especially in the context of EE.
West syndrome is a prototype of severe epileptic encephalopathy of infancy consisting of tonic spasms, psychomotor developmental regression, and the characteristic EEG pattern of hypsarrhythmia—a low structured mix of a multifocal spike and sharp wave activity and high voltage slow waves (Dulac, 2001a,b; Hrachovy & Frost, 2003). Early positron emission tomography (PET) studies have revealed a network underlying hypsarrhythmia that has included brain areas of focal cortical hypometabolism and subcortical hypermetabolism in the putamen and brainstem (Chugani et al., 1992; Chiron et al., 1993; Metsähonkala et al., 2002). It was unclear, however, whether this network is associated with epileptiform discharges or high amplitude slow activity, as both pathologic phenomena may be epileptogenic in West syndrome. In an electroencephalography–functional magnetic resonance imaging (EEG-fMRI) study, Siniatchkin et al. (2007) investigated infants with West syndrome and showed that the epileptiform discharges cause positive blood oxygen level–dependent (BOLD) signal changes in the cerebral cortex (especially in the occipital areas), whereas the high-amplitude slow wave activity in hypsarrhythmia is commonly associated with BOLD signal changes in brainstem, putamen, and thalamus (Fig. 1). There were no negative BOLD signal changes in relation to the hypsarrhythmia, especially no deactivations in the DMN. Moreover, the activation pattern seems to be specific for West syndrome, as it was not observed in infants with focal epilepsies. It is interesting to note that the subsequent source analysis, which was performed in the same sample of subjects for high amplitude slow activity, revealed similar electrical sources in the occipital cortex, putamen, and brainstem as shown by fMRI and even hypothesized a functional hierarchy between the sources, where the activity in brainstem seems to play an important key role in the pathogenesis of West syndrome (Japaridze et al., 2013). These data suggest that the involvement of brainstem is necessary and explains both the clinical seizures that might result from intermittent interference of descending brainstem pathways controlling spinal reflex activity and resulting in bilateral abrupt seizures, and the characteristic EEG features of hypsarrhythmia, which might be related to the activity in the ascending pathways from the same brainstem areas that project widely to the cerebral cortex (Juhasz et al., 2001; Frost and Hrachovy,2005). So, as suggested for West syndrome (Capovilla et al., 2009; Avanzini et al., 2012), the electroclinical phenotype could originate from the abnormal interaction of cortical and subcortical circuits rather than from a specific region alone.
It seems likely that the brainstem is a mandatory component to create the pathologic system responsible for both West syndrome and Lennox-Gastaut syndrome (LGS). West syndrome often evolves into LGS, an epileptic encephalopathy characterized by different types of seizures (tonic, tonic–clonic, and atonic seizures as well as atypical absences), typical EEG changes (slow spike-wave complexes ranging from 1 to 2.5/s, runs of rapid spikes and polyspikes), and accompanying mental deterioration (Arzimanoglou et al., 2009). An EEG-fMRI study in children with both symptomatic and cryptogenic cases of LGS revealed activation of brainstem and thalamus (Siniatchkin et al., 2011; see also Figs. 1 and 2) associated with interictal epileptiform discharges (IEDs). Brainstem plays a substantial role in mechanisms of tonic axial seizures, an important feature of the LGS: tonic-automatic seizures have been observed in a hydranencephalic patient (Velasco et al., 1997); tonic seizures fail to respond to callosotomy (Rougier et al., 1997); and tonic seizures are less responsive to resection of apparently epileptogenic cortical areas (Bladin, 1985). It has been hypothesized that clinical seizures might result from intermittent interference of descending brainstem pathways controlling spinal reflex activity and tonic neuronal bursting, whereas activity in the ascending pathways from the same brainstem areas that project broadly to the cerebral cortex might produce the characteristic EEG features in LGS (Blume, 2001). Because the brainstem exerts control over the gating function of the thalamus through its influence on the reticular thalamic nucleus, activity in brainstem (reticular formation and raphe serotoninergic pathways) and thalamus may lead to diffuse changes of cortical excitability that predispose a hyperexcitable neocortex to multifocal epileptic activity (Hrachovy & Frost, 2003). Surprisingly, a recently published study described an increased functional connectivity between brain areas belonging to the DMN in children with multifocal epileptic activity including young patients with the LGS (Moehring et al., 2013; Moehring J, Kroeher B, Galka A, von Ondarza G, Moeller F, Wolff S, Granert O, Jansen O, Boor R, Stephani U, Siniatchkin M, unpublished manuscript). This increased functional connectivity shows that the brains of patients with LGS are prone to increased synchrony, thereby predisposing them to multifocal epileptic activity. In addition, the study of Siniatchkin et al. (2011) demonstrated that especially the centromedian and anterior part of the thalamus are parts of the specific LGS system. A significant role of thalamocortical pathways in LGS was demonstrated previously: Centromedian thalamic nucleus stimulation in patients with refractory LGS has been an effective treatment strategy (Velasco et al., 2006). The anterior thalamus has been frequently and successfully used as a target for deep brain stimulation in patients with multifocal and secondary generalized epileptic activity (Samadani & Baltuch, 2007). These therapeutic options underscore the significance of the thalamus in the pathogenesis of LGS and multifocal IEDs.
Continuous Spikes and Waves during Slow Sleep
Although the thalamus has been found as a part of epileptic network in the EE with continuous spike and waves during slow sleep (CSWS) (Siniatchkin et al., 2010), thalamic activation is nonspecific in this epileptic condition and may explain EEG features of multifocality and a tendency for generalization in CSWS. The CSWS is an age-related disorder characterized by the presence of interictal epileptiform discharges during at least >85% of sleep and cognitive deficits associated with this EEG pattern. Despite etiologic heterogeneity, patients with CSWS were characterized by activation of a similar neuronal network—perisylvian region, insula, and cingulate gyrus, as well as deactivation in the DMN in all patients (see Figs. 1 and 3 as well as De Tiege et al., 2007; Siniatchkin et al., 2010). A comparison with the electrical source analysis results suggested that the activations corresponded to both initiation and propagation pathways of epileptiform discharges in the CSWS. Consistent with the study of Siniatchkin et al. (2010), other authors have performed magnetoencephalography (MEG)/EEG source analysis and have shown that the bilateral spikes generate in or propagate to the perisylvian cortex in many patients with Landau-Kleffner syndrome (LKS) and CSWS (Morrell et al., 1995; Sobel et al., 2000; Paetau, 2009). PET and single-photon emission computed photography (SPECT) studies have revealed brain areas of hypermetabolism and hypoperfusion in the perisylvian region and temporoparietal cortex associated with CSWS with and without LKS (Gaggero et al., 1995; Maquet et al., 1995; De Tiege et al., 2004; Luat et al., 2005; De Tiege et al., 2008). Although the perisylvian region seems to play an important role in the generation of spikes in the clinical tandem of LKS and CSWS, in patients with CSWS without LKS the generator may be located in other cortical areas and the epileptic activity seems to propagate bilaterally to the perisylvian cortex (Gaggero et al., 1995; Maquet et al., 1995; Morrell et al., 1995; De Tiege et al., 2004; Luat et al., 2005). Partial motor seizures that originate from different cortical regions (Tassinari et al., 2005) and focal cortical areas with hypermetabolism and hyperperfusion individually distributed and well corresponding with the focus of spikes during wakefulness (Gaggero et al., 1995; Maquet et al., 1995) provide evidence for individual cortical generators in CSWS. It is worthwhile to mention that this specific pattern of propagation in CSWS associated with bilateral synchrony in the perisylvian region is specific to CSWS rather than any etiologic factor. The described deactivations in structures of the DMN (Figs. 1 and 3) are consistent with the concept of epileptiform activity impacting on normal brain function by inducing repetitive interruptions of neurophysiologic function (Gotman et al., 2005; Laufs et al., 2007). Arguments for this suggestion provided De Tiege et al. (2008): Using longitudinal PET scans acquired before and after successful treatment of CSWS, the authors demonstrated common resolution of DMN hypometabolism associated with recovery. The authors hypothesized that the neurophysiologic effects associated with CSWS activity are not restricted to the epileptic focus but spread to connected brain areas via a possible mechanism of surrounding and remote inhibition.
Dravet syndrome or severe myoclonic epilepsy of infancy (SMEI) is an intractable epileptic encephalopathy of the early childhood that is caused by a mutation in the SCN1A gene in 80% of patients. In a recent EEG-fMRI study, 10 carriers of mutations in the SCN1A gene were investigated (Moehring et al., 2013; Moehring J, Kroeher B, Galka A, von Ondarza G, Moeller F, Wolff S, Granert O, Jansen O, Boor R, Stephani U, Siniatchkin M, unpublished manuscript). However, despite the common etiology, the study revealed different individual neuronal networks underlying interictal epileptiform discharges. The only common feature of brain activation consisted of positive BOLD signal changes in the DMN areas found in seven patients. But even these activations were inconsistent and did not contribute to the group analysis. Although the attempt to describe a common syndrome-specific network for SMEI patients was not successful, the results correspond with other neuroimaging studies that have demonstrated a pathogenetic heterogeneity in SMEI patients (Ferrie et al., 1997; Korinthenberg et al., 2004).
This review summarizes studies on functional neuroimaging that aimed to describe specific neuronal networks underlying EE. Indeed, different epileptic syndromes and EEs were characterized by specific fingerprints of brain activation (see Fig. 1). However, there is still insufficient explanation for cognitive deficits associated with epileptic activity in the context of EE. The interruption of activity in the DMN through epileptiform discharges and disturbed functional connectivity in the DMN provide one possible explanation. There are some EEs, however, that were not associated with the pathology in the DMN despite pronounced developmental abnormalities and cognitive deficits (see, for example, West syndrome, LGS, Dravet syndrome, and so on). Moreover, only few studies have investigated the relation between cognitive abilities, epileptiform discharges, and structure as well as function of neuronal networks directly by immediate testing of neuropsychological functions. Although the present studies enable one to build hypotheses about mechanisms of EE, more neuroimaging studies are needed to investigate short-term and long-term effects of epileptic activity on cognition and cognitive networks.
None of the authors has any conflict of interest to disclose. We confirm that we have read the Journal's position on issues involved in ethical publication and affirm that this report is consistent with those guidelines.