The Network or System‐oriented Approach in Understanding Human Epilepsy (Comments on the article of Susan S. Spencer: “Neural Networks in Human Epilepsy: Evidence of and Implications for Treatment”)

To the Editor: We read with great pleasure the work of Susan S. Spencer on the role of neural networks in human epilepsy. We join her approach by broadening the scope of the concept with further evidence. How these networks are related to the physiologic systems of the brain is an essential question. The author defines network as a “functionally and anatomically connected, bilaterally represented set of cortical and subcortical brain structures and regions in which activity in any part affects activity in all the others.” This definition fits nicely with different physiologic brain systems working essentially in functional networks and is the natural substrate of any epileptic disorder. From another aspect, the study of higher neural activity in several experimental and clinical paradigms leads to an important change from the concept of “one function, one structure” to the concept of “parallel distributed processing”(1). According to this principle, the higher cortical functions are represented in a network of several modules, and the same modules participate in different combination in different functions. The network can be influenced by affecting any of the contributing modules. We propose that the relation of these networks and the brain systems serving physiologic functions should be specified. The work enumerates some important epilepsy networks that also are well-known physiologic systems; however, some other well-known networks/systems harbor characteristic epileptic disorders. For example, the thalamocortical system, the epileptic disorder that is idiopathic generalized epilepsy, or the functional system connecting the mesiobasofrontal cortex and probably the basal ganglia, may play a role in hyperkinetic nocturnal forms of frontal lobe epilepsies. Presumably the network covering the frontoparietal parasagittal convexitiy and supracallosal medial surface also belongs to these complex systems. Here the common final pathway of the systems goes to the somatosensory supplementary motor (SSMA) area, and seizures result, with initial somatosensory symptoms evolving to the typical SSMA-type seizures. Another network having esssential importance in childhood is the perysylvian network, the epileptic disorder that is reflected in the benign focal childhood epilepsies and more malignant variants as Landau–Kleffner syndrome (LKS) and electrical status epilepticus in sleep (ESES). Some of these epileptic network abnormalities are heavily intermingled with developmental aspects. The early epileptic disorders necessarily modify the development of the affected system according to the principle of “firing and wiring together.” The work of Maquet et al. (2) on ESES and the recent pathophysiologic hypothesis about the origins of Lennox–Gastaut syndrome (LGS) by Blume (3) were born just in that spirit. The latter works show how epilepsy in the immature brain tends to transform pathologically the functional borders of the physiologic systems. Nevertheless, even these encephalopathic forms of epilepsy seem to appear within the framework of a developing phylogenetic cortical association system. The system/network approach among other advances might bridge over the contradictions in the concept of generalized and focal epilepsies. According to this concept, generalized epilepsies are the epileptic disorders of the midline thalamocortical systems with widespread bilateral projections. There is no reason either to call them generalized, or to distinguish them merely because bilaterality versus the predominant unilaterality, experienced in other epileptic networks, is more emphasized. Naturally, for surgical approaches, it is an important practical aspect, but theoretically, it should not make a difference. Here we wish only to emphasize the possibility of extending the network concept into a functional system-oriented approach in epilepsy categorization while discussion on reformatting the international classification of epilepsies exists.