Interictal EEG

Electroencephalography is the most specific method to define the epileptogenic cortex. Interictal epileptiform discharges, particularly if consistent over time, can provide useful information [6]. In temporal lobe epilepsy, consistently unitemporal interictal epileptiform discharges (IED) have a better prognosis for seizure freedom than bilateral IEDs. Focal, particularly extratemporal, epilepsies in which the EEG shows active regional polyspikes are more likely associated with cortical dysplasia as etiology of the epilepsy than patients with other IEDs [7]. Rhythmic midline theta activity, which is distinct from patterns of drowsiness of mental activation, is highly signficant for frontal lobe epilepsy and rarely seen in temporal lobe epilepsy [8]. This is particularly interesting since one out of four of these frontal lobe epilepsy patients did not show any interictal epileptiform discharges on noninvasive long-term EEG-monitoring and the rhythmic midline theta was the only interictal EEG abnormality [8].

Ictal EEG–video monitoring

Ictal EEG–video recording is critical in localizing the epileptogenic zone. A careful analysis of the first clinical signs and symptoms of a seizure and of the evolution of the seizure symptomatology can provide important clues on the epileptogenic zone [9–11]. One must keep in mind, however, that often an epileptic seizure arises from a “silent” region of cortex and would remain asymptomatic unless it spreads to “eloquent” cortex such as primary motor, primary sensory, or supplementary sensorimotor areas (Figure 2.1). Unfortunately, ictal EEG frequently documents discrepant results in extratemporal epilepsies [5]. Good concordance to MRI lesions and interictal EEG is only typical for temporal lobe epilepsy [5].

Figure 2.1 Illustration of the relation of the seizure onset and symptomatogenic zones. Seizure onset in the prefrontal region is likely to stay unnoticed unless the epileptic activity spreads into symptomatogenic cortex:

1. 
   

   1. Spread into the supplementary sensorimotor area leads to bilateral asymmetric tonic seizure

   
   
2. 
   

   2. Spread into the somatosensory hand area leads to right face clonic seizure

   
   
3. 
   

   3. Spread into the frontal eye field leads to right versive seizure

   
   
4. 
   

   4. Spread into frontal speech area leads to aphasic seizure

Seizure onset in the left occipital lobe leads to the following seizure evolution:

1. 
   

   5. Right visual aura ⇒ right versive seizure

Seizure onset in the temporal lobe leads to the following seizure evolution:

1. 
   

   6. Acoustic aura/abdominal aura ⇒ automotor seiure ⇒ right face clonic seizure.

Seizure semiology

Careful clinical observations and detailed reports of seizure semiology by the patient or observers have been used since the 18th century to classify epileptic seizures and epileptic syndromes. A detailed analysis of seizure semiology is still essential for the proper management of epileptic patients. A clear definition of the seizure type is important for classifying the epilepsy syndrome of the patient. The syndrome, together with the etiology of the epilepsy, are the essential factors determining the prognosis as well as the most effective pharmacological treatment. Seizure semiology plays an important role in the presurgical work-up, particularly when analyzed independently of other presurgical tests (EEG monitoring, neuroradiology, etc.). In addition, seizure semiology can be used effectively to differentiate between epileptic and nonepileptic seizures.

It is very important to emphasize that as a rule epileptic discharges limited to the seizure onset zone do not cause clinical symptoms unless located in an eloquent area (Figure 2.1). This is because the epileptogenic zone does not necessarily overlap with the symptomatogenic zone [1]. The term symptomatogenic zone refers to the area of the cortex that produces certain clinical symptoms as a result of epileptic activation. For example, seizures that originate in the frontal convexity remain asymptomatic as long as they do not spread into the symptomatogenic zones. If the epileptic activation reaches the primary motor area, versive or focal clonic seizures result (Figure 2.1). If the supplementary sensorimotor area is activated, focal tonic or hypermotor seizures occur; and if the activation spreads into the limbic system (cingulate gyrus), features of the seizure possibly become those of automotor seizures [12] (Figure 2.1). There is some association of specific seizure types with brain regions: seizures characterized by oral and manual automatisms (automotor seizures) [13, 14] are more common in temporal lobe epilepsy than in extratemporal epilepsy [15]. However, the specificity to temporal lobe epilepsy is much higher if automotor seizures are preceded by epigastric (abdominal) auras [15]. Similarly, unilateral clonic seizures of the face are frequently seen in patients with paracentral epilepsies. However, the same seizure type may occur in patients with temporal lobe epilepsy but then is usually preceded by manual and oral automatisms (automotor seizure. In fact, this evolution is more likely to occur in lateral than mesial temporal lobe epilepsy [16]. Unilateral clonic seizures may be associated with both frontal and temporal lobe epilepsies. However, the sequence of the seizure evolution makes a major difference. In temporal lobe epilepsy, unilateral facial clonic seizures are typically preceded by manual and oral automatisms, which are rarely the case in frontal lobe epilepsy. Thus, the association of single-seizure types to particular localizations of the epileptogenic zones is not as strong as the association of the evolution of seizure types to specific brain regions. This may explain why several studies which neglected this fact found poor localizing value of seizure semiology [17]. Another limitation is that many studies relied on the description of seizures rather than investigating adequate video-recorded seizures [17]. Patients’ or witnessess’ descriptions of seizure are subject to bias and not sufficiently reliable.

Table 2.1 summarizes studies on nonlesional epilepsy patients. A computerized online search via MEDLINE (online PubMed from first available year to April 2013) using the search term “nonlesional epilepsy” identified 121 studies, of which 78 were excluded for being review articles (n = 16), meta-analysis (n = 1), or a commentary (n = 1). Animal studies (n = 2), genetic studies (n = 4), as well as ten studies investigating symptomatic epilepsy and 19 not clearly differentiating between nonlesional and lesional epilepsy patients (n = 19) were excluded. Seven publications including patients with status or generalized epilepsy syndromes were excluded, as they do not localize in early infancy. In addition, studies were excluded if they did not use MR imaging (n = 6) or were written in languages other than English or German (n = 9). Two publications were excluded, because the full text was not available online. In total, 43 studies met our inclusion criteria.

Table 2.1 Characteristics of selected studies about nonlesional epilepsy

n = number of subjects, m = male, f = female, bihem. = bihemispheric, histop. = histopathology, r = retrospective, p = prospective, UCT = uncontrolled clinical trial, CCT = controlled clinical trial, CR = case report, SPS = simple partial seizure, CPS = complex partial seizure, GTC = generalized tonic–clonic seizure, SGC = secondary generalized seizure, n.a. = no data available, a. = data available, TL = temporal lobe, FL = frontal lobe, PL = parietal lobe, OL = occipital lobe, CR = central region, Y = yes, N = no.

Different terminologies were used to classify or label seizure semiology. It was mostly labeled after the lobe, such as temporal lobe seizure or frontal lobe seizure, providing no reliable clinical information on the seizure characteristics. Other studies used the seizure classification system of the International League against Epilepsy with terms such as complex partial seizures (CPS) or simple partial seizures (SPS). These terms only provide the information whether consciousness is disturbed or not in patients with focal epilepsies regardless of the actual seizure semiology. Only few publications of case series provide detailed information on seizure semiology and EEG findings [18–20]. We, therefore, use the semiological seizure classification to provide clinically localizing information [13, 14]. With the help of EEG–video-recorded seizures, several very reliable lateralizing signs have been identified which have an accuracy of 80–100% (Table 2.2) [12, 21].

Table 2.2 Lateralizing seizure phenomena

The EEG data were mostly reported as being concordant or discordant with the other diagnostic findings (Table 2.1). Highest diagnostic sensitivity in the localization of epileptogenic foci and seizure lateralization was demonstrated for ictal scalp EEG. Concordance rate was higher in the good than in the poor surgical outcome group [22].

The lateralizing and localization value of seizure semiology, and their role in MRI-negative surgery, are further discussed elsewhere in this book according to the brain regions affected by epilepsy (Chapters 14, 15, 16, and 18), in children (Chapter 17), and with relevance to cortical mapping (Chapter 13).

Illustrative patients

How seizure semiology and EEG help to develop a hypothesis on the epileptogenic zone in patients with negative MRI is illustrated by the following two patients: