Analysis of the initial ictal phenomenon in patients with temporal lobe epilepsy
Article Outline
- Abstract
- 1. Introduction
- 2. Materials and methods
- 3. Results
- 4. Discussion
- Conflicts of interest statement
- References
- Copyright
Abstract
We aimed to assess the localizing value of the initial semiological element in temporal lobe epilepsy (TLE). Video-EEG-documented seizures of 97 adult TLE patients were studied in relation to seizure origin (left versus right; mesial versus extra-mesial). Strikingly, seizures with mesial onset started with very few ictal phenomena, while seizures of extra-mesial origin began with a larger variety of ictal elements. Furthermore, following noticeable distributions were observed for the mesial group: (i) aura was the most common initial ictal phenomenon in the total patient collective, occurring significantly more frequently in mesial than in extra-mesial seizure onset. Aura appeared most often in seizures of left mesial origin. (ii) Vocalization presented a trend towards mesial left seizure origin. (iii) Oral automatisms showed a trend towards mesial seizure origin. Following noticeable distribution was observed for the extra-mesial group: In patients without aura, restlessness as initial ictal phenomenon appeared exclusively in seizures of extra-mesial right origin. Finally, behavioral arrest showed a trend towards left-sided seizure origin. In conclusion, the initial ictal element may add useful information concerning differentiation of seizure onset in TLE.
Keywords: Chronological, First ictal phenomenon, Ictal semiology, Mesial, Extra-mesial, Temporal lobe epilepsy
1. Introduction
The success of surgical treatment of TLE is based upon a close selection of candidates with consistent presurgical diagnostic findings in seizure semiology, electrophysiology, neuropsychology, and neuro-imaging of the individual patient. In this context, the analysis of ictal semiology is crucial, providing additional information on lateralization and localization of seizure onset.1 A number of semiological elements, such as behavioral arrest, restlessness, and staring, are frequently observed as features of temporal lobe seizures, but the point in time of their occurrence within the entire seizure course has not been systematically investigated. Only auras are generally known to occur as first seizure phenomena in TLE. Yet, the analysis of the initial ictal element is of particular interest as it may be an early indicator of the localization of seizure onset. In the present study, we analyzed video-EEG-documented seizures of adult patients with refractory TLE of mesial or extra-mesial origin with respect to the initial ictal phenomenon during the seizure course. We aimed to determine whether the characteristic of the initial semiological element can indicate the localization and/or lateralization of seizure origin.
2. Materials and methods
The study was performed in accordance with the Declaration of Helsinki and approved by the local ethics committee. All patients gave their informed consent prior to inclusion in the study. We analyzed 97 surgically treated adult patients with medically refractory TLE of mesial or extra-mesial seizure origin (for clinical data see Table 1, for histological data see Table 2). All patients had complete presurgical evaluation, including medical, neurological and neuropsychological examination, video-EEG monitoring, and cranial MRI, and were postoperatively seizure-free for at least two years after epilepsy surgery. From each of these TLE patients, the first presurgical video-EEG-documented seizure was analyzed for the succession of semiological elements; every single seizure phenomenon was listed and numbered chronologically. The analysis was performed by two independent investigators, blinded to clinical details and to the results of the monitoring, in the following steps (Table 3): First, we studied the total collective for the initial ictal element after seizure onset. Then, the total collective was divided into two subgroups: (1) the group of patients with aura, in which the first objective seizure element was determined; (2) the group of patients without aura, in which the initial ictal phenomenon was documented. In all groups, the frequency of occurrence of the respective ictal elements was dissected with regard to seizure origin (left versus right; mesial versus extra-mesial). Statistical analysis was performed using SPSS. Frequency tables were analyzed by the log-linear model.
Table 1. Clinical data of patients.
| Mesial seizure origin | Extra-mesial seizure origin | |
|---|---|---|
| Number of patients | Left mesial: n =35 | Left extra-mesial: n=18 |
| Right mesial: n=27 | Right extra-mesial: n=17 | |
| Total: n=62 | Total: n=35 | |
| Sex | 29 F, 33 M | 19 F, 16 M |
| Mean age±SEM | 37±7 years | 33±7 years |
Table 2. Histological data of patients.
| Seizure origin | Histology | Number of patients |
|---|---|---|
| Mesial | Ammon's horn sclerosis | Left: n=35a |
| Right: n=27a | ||
| Total: n=62a | ||
| Extra-mesial | Hamartoma | Left: n=4b |
| Right: n=5b | ||
| Total: n=9b | ||
| Extra-mesial | Astrocytoma | Left: n=5b |
| Right: n=4b | ||
| Total: n=9b | ||
| Extra-mesial | Ganglioglioma | Left: n=9b |
| Right: n=8b | ||
| Total: n=17b | ||
Table 3. Number of patients with mesial and extra-mesial TLE.
| Mesial left | Mesial right | Extra-mesial left | Extra-mesial right |
|---|---|---|---|
| (A) Total patient collective | |||
| 35 | 27 | 18 | 17 |
| (B) Patients with aura | |||
| 91.4% (32/35) | 81.5% (22/27) | 61.1% (11/18) | 52.9% (9/17) |
| (C) Patients without aura | |||
| 8.6% (3/35) | 18.5% (5/27) | 38.9% (7/18) | 47.1% (8/17) |
3. Results
Fig. 1 shows a summary of the results for the initial ictal element in the total patient collective, Fig. 2 describes the findings for the first objective seizure phenomenon in the group of patients with aura, Fig. 3 presents the results for the first ictal element in the group of patients without aura. Considering the total collective of patients, it is striking that seizures with mesial onset start with only few ictal phenomena as first seizure elements (Fig. 1). Seizures with left mesial onset begin with either aura, behavioral arrest, staring, or dystonic posturing. Seizures with right mesial onset start with either aura, oral automatisms, stereotyped movements, or head turning to the ipsilateral side. In contrast, seizures with extra-mesial onset begin with a larger variety of ictal phenomena as first seizure elements (Fig. 1).
Fig. 1.
Analysis of initial ictal phenomena in the total collective of patients. (A) Seizures of mesial origin. (B) Seizures of extra-mesial origin. Ictal elements are vertically arranged with regard to decreasing frequency of occurrence. a–d: Ictal phenomena marked by the same letter occur with identical frequency.
Fig. 2.
Analysis of the first objective ictal phenomena in the group of patients with aura. (A) Seizures of mesial origin. (B) Seizures of extra-mesial origin. Ictal elements are vertically arranged with regard to decreasing frequency of occurrence. a–f: Ictal phenomena marked by the same letter occur with identical frequency.
Fig. 3.
Analysis of initial ictal phenomena in patients without aura. (A) Seizures of mesial origin. (B) Seizures of extra-mesial origin. Ictal elements are vertically arranged with regard to decreasing frequency of occurrence. a–d: Ictal phenomena marked by the same letter occur with identical frequency.
Moreover, noticeable distributions were observed for the ictal phenomena aura, vocalization, oral automatisms, restlessness, and behavioral arrest, and are specified below.
3.1. Initial ictal elements associated with seizures of mesial origin
Table 4. Relevant initial ictal phenomena for TLE of mesial origin.
Mesial left Mesial right Extra-mesial left Extra-mesial right (A) Aura as initial ictal phenomenon in the total collective of patients 91.4%
(32/35)81.5%
(22/27)61.1%
(11/18)52.9%
(9/17)(B) Vocalization as first objective ictal phenomenon in the group of patients with aura 15.6%
(5/32)9.1%
(2/22)0.0%
(0/11)11.1%
(1/9)(C) Oral automatisms as first objective ictal phenomenon in the group of patients with aura 12.5%
(4/32)22.7%
(5/22)0.0%
(5/22)11.1%
(1/9)
3.2. Initial ictal element associated with seizures of extra-mesial origin: restlessness
Regarding the group of patients with aura and focussing on the first objective seizure element, a wide variety of ictal phenomena was observed (Fig. 2). Conspicuously, restlessness constituted the most common first objective ictal element in all subgroups (mesial left: 21.9%, mesial right: 27.3%, extra-mesial left: 36.4%, and extra-mesial right: 55.6%; Table 5A, Fig. 2). Thus, the most frequent succession of ictal elements in temporal lobe seizures of mesial as well as of extra-mesial origin was constituted by aura as initial seizure phenomenon, accompanied by restlessness as first objective seizure element. This succession of seizure phenomena showed a preference for the group of extra-mesial right seizure onset. Here, statistical analysis did not reveal significant group differences. Notably, in the patient sample without aura, restlessness appeared exclusively in seizures of extra-mesial right origin (Table 5B; Fig. 3).
Table 5. Relevant initial ictal phenomena for TLE of extra-mesial origin.
| Mesial left | Mesial right | Extra-mesial left | Extra-mesial right |
|---|---|---|---|
| (A) Restlessness as first objective ictal phenomenon in the group of patients with aura | |||
| 21.9% (7/32) | 27.3% (6/22) | 36.4% (4/11) | 55.6% (5/9) |
| (B) Restlessness as initial ictal phenomenon in the group of patients without aura | |||
| 0.0% (0/3) | 0.0% (0/5) | 0.0% (0/7) | 25.0% (2/8) |
3.3. Initial ictal element associated with seizures of left temporal origin: behavioral arrest
With regard to the total collective of patients, behavioral arrest occurred only rarely as initial ictal element. However, when focussing on the first objective seizure phenomenon in the group of patients with aura, we found a trend towards left-sided temporal seizure origin (mesial left: 21.9% and extra-mesial left: 27.3%; versus mesial right: 4.5% and extra-mesial right: 11.1%; Table 6). Statistical analysis did not reveal significant group differences.
Table 6. Relevant initial ictal phenomena for TLE of left-sided origin.
| Mesial left | Mesial right | Extra-mesial left | Extra-mesial right |
|---|---|---|---|
| (A) Behavioral arrest as first objective ictal phenomenon in the group of patients with aura | |||
| 21.9% | 4.5% | 27.3% | 11.1% |
| (7/32) | (1/22) | (3/11) | (1/9) |
4. Discussion
Ictal semiology is essential for the clinical diagnosis of epilepsy. In addition, it is important for the evaluation of patients for epilepsy surgery, as it may provide further information about the site of seizure onset.1 A number of semiological features have been described as typical for TLE. Among these, the most frequent phenomena include aura, behavioral arrest, and oro-alimentary automatisms.2, 3, 4, 5, 6 However, the point in time of their occurrence within the entire seizure course has not been systematically studied to date. Actually, the nature of the initial ictal element is of special interest as it might be an early indicator of the localization of seizure origin.
In our present study, we investigated the initial ictal phenomenon in seizures of adult TLE patients in relation to the epileptogenic origin (mesial versus extra-mesial; left versus right). Our analysis showed a striking difference between seizures of mesial and seizures of extra-mesial onset. Seizures of mesial origin started with only few ictal phenomena. In the left mesial group, they began with either aura, behavioral arrest, staring, or dystonic posturing. In the right mesial group, seizures started with either aura, oral automatisms, stereotyped movements, or head turning to the ipsilateral side. In contrast, seizures of extra-mesial origin began with a much larger variety of initial ictal phenomena. However, we could not identify any semiological element that would solely and significantly correlate with a single localization of seizure onset. Still, the nature of the initial seizure phenomenon may give a first implication towards localization of seizure onset. In this context, several notable ictal elements have to be discussed.
Analysis of the total patient collective showed that aura constituted the most common initial ictal phenomenon, occurring significantly more frequently in mesial than in extra-mesial seizure onset. In addition, aura presented most often in seizures of left mesial origin. Thus, the occurrence of aura may provide a first hint at localization of seizure focus.
In a further step, we focussed on the group of patients with aura, analyzing the first objective ictal phenomenon. Interestingly, vocalization never presented as initial seizure element in the total patient collective, but occurred exclusively as first objective seizure element in the group of patients with aura, with a trend towards mesial left seizure origin. Thus, the combined occurrence of aura together with vocalization as first objective seizure element may give an implication towards mesial left localization of seizure focus. Notably, all these patients with vocalization as first objective seizure element were right-handed with language dominance on the left side. Thus, there was no correlation between side of language dominance and side of seizure onset. This corresponded well to an earlier study by Morrell et al.,7 showing that seizures with spontaneous vocalization were as likely to originate from language-dominant as from language-nondominant temporal cortex.
Oral automatisms occurred only rarely as first ictal element in the total patient collective, but focussing on the first objective seizure element in the group of patients with aura, we found a trend towards mesial seizure origin. Likewise, Gil-Nagel and Risinger3 found that early oral automatisms occurred significantly more frequently in hippocampal temporal lobe seizures, when compared to extra-hippocampal temporal lobe seizures. Thus, the combined occurrence of aura together with oral automatisms as first objective seizure element may give a stronger hint at mesial localization of seizure focus.
Similarly, behavioral arrest was only rarely found as initial ictal element in the total patient collective, but regarding the first objective seizure element in the group of patients with aura, there was a trend towards left-sided seizure origin. This complied with our earlier study of behavioral arrest in TLE,8 which showed that behavioral arrest was observed more frequently in left-sided temporal seizures. This earlier study analyzed the occurrence of behavioral arrest in general and did not focus on behavioral arrest as initial seizure element. Nevertheless, we then noted that behavioral arrest within the sequence of seizure elements was remarkably consistent, being observed early at seizure onset. In addition, Steinhoff et al.4 also found movement arrest at seizure onset being typical of left TLE. Taken together, the occurrence of aura together with behavioral arrest as first objective seizure element may give a combined implication towards mesial left localization of seizure focus.
Finally, we studied the group of patients without aura for the initial ictal phenomenon after seizure onset. Here, restlessness appeared exclusively in seizures of extra-mesial right origin. Thus, the occurrence of restlessness as initial ictal element may give a strong hint at extra-mesial right localization of seizure focus.
In our present study, we analyzed the respective first presurgical video-EEG-documented seizure of 97 surgically treated TLE patients. The total number of documented seizures of these patients is higher, because for each patient, one to three seizures were video-EEG-recorded. In order to test, whether this point would influence the validity of the results described above, we additionally performed the complete analysis for the extended number of seizures, including all available video-EEG-documented seizures of these 97 TLE patients. Strikingly, intra-individual semiology of the very first seizure elements was remarkably constant in the majority of patients. Consequently, the results as described in this paper remained valid.
Furthermore, it should be noted, that during the presurgical video-EEG-monitoring, all patients of this study were not on their usual antiepileptic medication, but in the process of reduction of antiepileptic medication. This fact has to be kept in mind when applying the results of this study, as antiepileptic drugs may alter semiology within an individual.
Unquestionably, the results of our study have to be regarded with caution, because – with the exception of aura – none of the other seizure elements reached statistically significant group effects. Yet, the special succession of the mentioned semiological features can be at least supportive for localization of seizure origin, especially when combined with other significant ictal phenomena and integrated with EEG and neuro-imaging data.
In conclusion, the determination of the initial seizure phenomenon, or the determination of the occurrence of aura in combination with the first objective seizure element, may add useful information concerning localization of seizure onset in TLE. Doubtlessly, seizure semiology should always be integrated with comprehensive EEG and neuro-imaging data; nevertheless, the acquisition of a valid set of semiological features with reliable localizing value is beneficial to define more precisely the epileptogenic zone and the subsequent surgical procedure.
Conflicts of interest statement
None of the authors has any conflicts of interest to disclose.
References
- . Semiology of temporal lobe seizures: value in lateralizing the seizure focus. Epilepsia. 1998;39:721–726
- . Psychomotor seizures of temporal lobe onset: analysis of symptom clusters and sequences. Epilepsy Research. 1995;20:49–67
- . Ictal semiology in hippocampal versus extrahippocampal temporal lobe epilepsy. Brain. 1997;120:183–192
- . The lateralizing value of ictal clinical symptoms in uniregional temporal lobe epilepsy. European Neurology. 1998;39:72–79
- . Localizing and lateralizing value of epileptic symptoms in temporal lobe epilepsy. Canadian Journal of Neurological Science. 2000;27:pp. 1-5; 20-21
- Clinical-electroencephalogram patterns at seizure onset in patients with hippocampal sclerosis. Clinical Neurophysiology. 2003;114:2286–2293
- . Speech during partial seizures: intracranial EEG correlates. Epilepsia. 1991;32:886–889
- . Lateralizing value of behavioral arrest in patients with temporal lobe epilepsy. Epilepsy & Behavior. 2008;13:634–636
PII: S1059-1311(10)00034-8
doi:10.1016/j.seizure.2010.02.005
© 2010 British Epilepsy Association. Published by Elsevier Inc. All rights reserved.
