If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Department of Pediatrics, Gyengsang National University School of Medicine, Jinju, Republic of KoreaGyeongsang Institute of Health Science, Jinju, Republic of Korea
Corresponding author at: Department of Neurology, Gyeongsang National University School of Medicine, 816-15 Jinjudae-ro, Jinju 660-751, Republic of Korea. Tel.: +82 55 750 8288; fax: +82 55 750 1709.
Gyeongsang Institute of Health Science, Jinju, Republic of KoreaDepartment of Neurology, Gyengsang National University School of Medicine, Jinju, Republic of Korea
We observed current-source of spikes in childhood absence epilepsy (CAE).
•
Frontal lobe was involved in all of the enrolled cases.
•
Temporal lobe involvement was associated with the initial-drug failure.
•
The information may help to estimate the efficacy of initial treatment in advance.
Abstract
Purpose
In addition to the frontal lobe, the temporal lobe may also be involved in typical absence seizures. However, few studies have addressed the relationship between this involvement and drug responsiveness in childhood absence epilepsy (CAE). In this study, we observed the current-source distribution (CSD) of generalized spike-and-wave discharges (GSWDs) and investigated the relationship between temporal lobe involvement in the CSD and responsiveness to initial antiepileptic drug (AED) in CAE.
Method
Seventeen consecutive patients with CAE were retrospectively enrolled in the study. Patients were divided into an initial-response group and an initial-failure group, according to their responsiveness to the initial AED treatment. For each patient, the spike peak CSD of an averaged GSWD was obtained from the initial electroencephalogram. We compared the incidence of temporal involvement in the CSD between the two groups. We also compared clinical variables, including age of onset, gender, type and dose of first AED, time to cessation of clinical seizures, and seizure-free status.
Results
The initial-response and initial-failure groups contained 12 and five patients, respectively. Temporal lobe involvement was more frequent (80% vs. 17%, p = 0.03), and time to cessation of clinical seizures was more prolonged (median 2.5 months vs. 8 months, p < 0.01) in the initial-failure than in the initial-response group. None of the other variables studied differed between groups.
Conclusion
Initial AED failure was associated with temporal involvement in the CSD of CAE patients. This electrophysiological information may be helpful in clinical practice by estimating the efficacy of initial AED treatment in AED-naïve CAE patients in advance.
in: Niedermeyer E. Lopes ds Silva F. Electroencephalography: basic principles, clinical applications, and related fields. 6th ed. Wolters Kluwer/Lippincott Williams & Wilkins Health,
Philadelphia2011: 479-540
Physiopathogenesis of generalized epilepsies of organic nature (stereoencephalographic study).
in: Gastaut H. Jasper H.H. Bancaud J. Waltregny A. The physiopathogenesis of the epilepsies. 3rd ed. Charles C Thoma Publisher,
Springfield1969: 158-185
in: Niedermeyer E. Lopes ds Silva F. Electroencephalography: basic principles, clinical applications, and related fields. 6th ed. Wolters Kluwer/Lippincott Williams & Wilkins Health,
Philadelphia2011: 479-540
Physiopathogenesis of generalized epilepsies of organic nature (stereoencephalographic study).
in: Gastaut H. Jasper H.H. Bancaud J. Waltregny A. The physiopathogenesis of the epilepsies. 3rd ed. Charles C Thoma Publisher,
Springfield1969: 158-185
]. These findings suggests that the frontal lobe are hyperexcitable, and have epileptogenicity related to absence seizures and/or GSWDs.
Although there are some common locations for epileptic cortices, individual variation in regional susceptibility of the cortex may lead to variability in the localization and size of generators for GSWDs [
in: Niedermeyer E. Lopes ds Silva F. Electroencephalography: basic principles, clinical applications, and related fields. 6th ed. Wolters Kluwer/Lippincott Williams & Wilkins Health,
Philadelphia2011: 479-540
Physiopathogenesis of generalized epilepsies of organic nature (stereoencephalographic study).
in: Gastaut H. Jasper H.H. Bancaud J. Waltregny A. The physiopathogenesis of the epilepsies. 3rd ed. Charles C Thoma Publisher,
Springfield1969: 158-185
]. Despite this, temporal lobe involvement in the generation of typical absence seizures has had little recent attention. Furthermore, certain clinical aspects of atypical absence seizures differ somewhat from those of typical absence seizures [
]. Atypical absence seizures, on the other hand, are invariably associated with severe symptomatic or cryptogenic epilepsies, and tend to be refractory to AED medication [
], we hypothesize that temporal involvement in seizure generation may be an indicator of drug refractoriness.
Although the long-term prognosis of CAE is excellent in general, 20–71% of patients with CAE experience treatment failure with the initial antiepileptic drugs [
]. The initial AED failure may therefore be an early indicator of the long-term prognosis for CAE patients. We hypothesized that temporal involvement in seizure generation may be associated with initial treatment failure in patients suffering from typical absence seizures. We investigated the relationship between temporal lobe involvement in the current-source distribution (CSD) of spikes and failure of the initial AED in CAE. To define the CSD of patients with CAE, we employed a distributed model of current-source analysis of GSWDs.
2. Methods
2.1 Subjects
Children who were newly diagnosed with epilepsy and who met the diagnostic criteria for CAE were recruited retrospectively from January 2008 to January 2014 at Gyeongsang National University Hospital and Samsung Changwon Hospital. This retrospective study was approved by the institutional review boards of both hospitals (GNUH-2014-12-028, SCMC-2015-016). The diagnosis of CAE was based on the criteria proposed by Panayiotopoulos and reviewed by Loiseau et al. [
] The criteria for inclusion were (1) seizure onset between 4 and 10 years; (2) normal neurological state and development; (3) absence seizures as the initial type of seizures; (4) very brief typical absence seizures occurring many times per day; and (5) epileptiform discharges of bilateral, symmetrical, and synchronous discharge of regular 3 Hz SWCs with normal or mildly abnormal background activity. The criteria for exclusion were: (1) atypical clinical features, such as generalized tonic clonic seizures or myoclonic jerks, before or during the active stage of absence, absences with marked eyelid or perioral myoclonus, stimulus-sensitive absences; and (2) atypical EEG features, such as discharge fragmentation and multiple spikes, irregular and multiple spike and slow-wave discharges with marked variations in the intradischarge frequency, predominant brief discharges of 3–4 Hz spike-waves of <4 s, or fixed ‘lead in’ anomaly in the frontal region on EEG.
During the study period, a total of 24 patients were newly diagnosed with CAE, according to the inclusion criteria. Among them, seven patients were excluded for the following reasons: (1) atypical clinical features, according to our exclusion criteria (n = 2); (2) atypical EEG features, according to our exclusion criteria (n = 2); (3) patient unavailable for EEG before AED medication (n = 1); (4) patient unavailable for drug responsiveness assessment (n = 2). Ultimately, 17 patients were enrolled in the study. The median age of onset was 8 years (range, 5–10 years), and the median treatment period was 3.0 years (range, 1.2–6.5 years). Patients were divided into two groups according to responsiveness to initial antiepileptic drug (AED) treatment: an initial-response group and an initial-failure group. Initial failure of AED treatment was defined as the inability to attain complete seizure control with the optimal dose of the first appropriate AED.
2.2 EEG recording and collection of average spikes
Scalp EEGs were recorded for a minimum of 30 min in each patient, according to the International 10–20 system, with 25 or 19 channels, depending on the clinical setting of each hospital. A 32-channel digital EEG machine (Comet® EEG machine; Grass-Telefactor; West Warwick, RI, USA) was used in each hospital. Sampling rates were 400 or 200 Hz, depending on the clinical setting. We analyzed an EEG recording taken before AED medication, and selected artifact-free EEG segments of 500 ms before and 500 ms after the spike peak points of typical 3 Hz GSWDs. We obtained an averaged spike of 18–50 GSWDs from each patient. Because the numbers of typical spikes varied in each EEG, the numbers of spikes averaged also varied. The filter setting was 1.6–30 Hz. Brain Electrical Source Analysis (BESA; V. 5.1; MEGIS; Grafelfing, Germany) software was used to select epochs for data processing.
2.3 eLORETA images and localization of the current-source distribution
Exact low-resolution brain electromagnetic tomography (eLORETA, Key Institute for Brain-Mind Research; Zürich, Switzerland), a functional image modality expressing current-source distribution (CSD) in three-dimensional brain images of 6239 voxels, with a space resolution of 5 mm [
], was used to obtain CSD images. eLORETA images corresponding to the positive peak of the averaged spike were obtained from each patient to identify the anatomical distribution and extent of the current-source of the averaged spike. To objectively define CSD, percentiles were used to determine the threshold of significance. Because the probability distribution for the current-source suggested that the 95th percentile would be suitable, a current-source distribution above the 95th percentile was designated as the threshold of significance.[
Changes in current-source density of interictal spikes in benign epilepsy of childhood with centrotemporal spikes following treatment with oxcarbazepine.
] Therefore, we calculated an eLORETA value of mean + 2 standard deviations among all eLORETA values in each EEG and considered this the threshold value of eLORETA images in each EEG. The neuronal cortex has been modeled as a collection of volume elements (voxels) in the digitized Talairach atlas (provided by the Brain Imaging Center, Montreal Neurological Institute). eLORETA analysis was performed on the complete EEG data by an investigator (SJ) who did not have access to any patient clinical information during the entire analytic process. The location of the CSD in each patient was identified from the eLORETA images.
2.4 Acquisition of clinical data
Clinical data were acquired by retrospective review of medical records. Clinical variables included age of onset, age at initial EEG recording, gender, type of initial AED, presence or absence of response to initial AED treatment, time interval to achievement of ultimate seizure control, drug dosage at the time of determination of the success or failure of the first AED, and seizure freedom. Successful response to initial AED was defined as complete seizure control with the first AED. Patients were divided into two groups according to responsiveness to initial AED treatment: the initial-response group and the initial-failure group. Initial failure of AED treatment was defined as a lack of complete seizure control with an optimal dose of the first appropriate AED. Seizure control was defined as the absence of clinical absence seizures. The clinical cessation of absence seizures was defined as a lack of seizures in parent documentation as well as failure to provoke seizures using a bedside hyperventilation induction test. Seizure freedom was defined as freedom from seizures during the preceding year under proper medication.
2.5 Statistical analysis
The clinical variables and the location of the CSD were compared between the two study groups using a Mann–Whitney U-test for numerical data and Fisher's exact test for categorical data; p-values < 0.05 were considered statistically significant. All analyses were performed using SPSS for Windows, version 12.0 (Chicago, IL, USA) (Table 1).
Table 1Comparison of clinical characteristics and current-source distribution between initial-responsive group and initial-failure group.
Among the 17 patients, 12 patients were responsive to the first AED, and the other five patients failed to respond to the first AED (Table 1). In the patients who were not responsive to the initial drug, an alternative AED was either substituted for or added to the first AED treatment regimen. The time interval to reach seizure control was a median of 2.5 months (range 1–7 months) in the initial-response group. In the initial-failure group, two patients did not reach ultimate seizure control, even after a treatment period of more than 5 years with add-on therapy. The median time interval to ultimate seizure control was 8 months (range, 6–16 months) in the remaining 15 patients. Seizure freedom was more common in the initial-response (12/12; 100%) than in the initial-failure group (3/5; 60%; p = 0.07).
3.2 Comparison of the current-source distribution of averaged spikes in the initial-response and initial-failure groups
The location and lateralization of involved areas are described in Table 2. In all cases, the current-source of averaged spikes was distributed in the frontal area of both cerebral hemispheres. The bilateral frontal involvement of the CSD was in 11 or 12 patients (91.7%) in the initial-response group (Fig. 1), and all 5 patients (100%) in the initial-failure group. Six of 17 patients had temporal lobe involvement in the current-source distribution (35.3%) (Fig. 2). Temporal lobe involvement was significantly different between the two groups (p = 0.03); 2 of 12 patients in the initial-response group (16.7%), whereas 4 of 5 patients in the initial-failure group (80.0%) (Table 1).
Table 2Comparison of current-source locations between initial-responsive group and initial-failure group.
Fig. 1Results for a 5-year-old patient with childhood absence epilepsy in the initial-response group. An averaged spike was obtained from 50 interictal generalized spike-and-wave discharges (GSWDs) (A). The current-source at the spike peak point of the averaged interictal GSWD was distributed within the frontal area of the both cerebral hemispheres (B).
Fig. 2Results for a 9-year-old patient with childhood absence epilepsy in the initial-failure group. An averaged spike was obtained from 31 interictal generalized spike-and-wave discharges (GSWDs) (A). The current-source at the spike peak point of the averaged interictal GSWD was distributed within the frontal and temporal areas of both cerebral hemispheres (B).
3.3 Comparison of clinical characteristics between the initial-response group and the initial-failure group
The clinical characteristics of the two groups are compared in Table 1. Age of onset, age at initial EEG recording, gender, type of first AED, treatment period, frontal lobe involvement in CSD, and seizure freedom did not differ between the two groups.
4. Discussion
In this study, we investigated the relationship between extrafrontal–temporal involvement in the CSD of spikes and responsiveness to initial AED treatment in CAE. Our current-source analysis of GSWDs using eLORETA revealed an association between temporal involvement in CSD and failure to respond to initial AED treatment. Age of onset, gender, type of first AED, seizure freedom, and frontal involvement in CSD did not differ between the two groups. The CSD in CAE may therefore differ between patients experiencing different seizure control states. These results are concordant with the emerging concept that the composition of the epileptic network may influence drug responsiveness in IGE and CAE [
]. This association between electrophysiology and drug responsiveness may help to estimate drug responsiveness in AED-naïve patients with CAE in advance.
To the best of our knowledge, no study has been conducted to date concerning an association between drug responsiveness and generator localization in CAE. In adult patients with IGE, a few studies have examined this issue using EEG and functional magnetic resonance imaging (fMRI) [
] showed that absence seizures in drug-resistant adult patients with IGE initiated from widespread cortical areas, including the prefrontal and dorsolateral cortices, before propagating to the thalamus. They also demonstrated that blood oxygen level dependent (BOLD)-effect limbic activation was strongest at the time of seizure onset, coinciding with activation of the temporal and thalamic areas [
]. Therefore, the cortical activation pattern observed at the time may not have been unique to drug-refractory IGE patients. A few years later, Szaflarski et al. [
] compared BOLD signals between drug-resistant and drug-responsive adults with IGE during GSWDs. They showed that the GSWD generators of drug-resistant patients featured more extensive cortical involvement compared with those of drug-responsive patients [
]. At the onset of GSWDs, the medial frontal cortex, paracingulate gyrus, and anterior insula were activated bilaterally in drug-resistant patients but not in drug-responsive patients [
]. Although various cortical regions, such as the frontal, parietal, and temporal lobes and the insula, may form the leading foci that generate GSWDs [
], most studies involving typical absence seizures have addressed the importance of the frontal lobe. A study using current-source analysis of EEG data showed that the frontopolar regions were typically associated with seizure onset and were the most activated cortical regions during propagation of seizures in five adult patients with absence epilepsy [
Although the frontal area is well known to be a critical area for the generation of GSWDs and absence seizures, little is known about the roles of the temporal lobe and the limbic structures as generators of typical absence seizures. The cortical areas activated by absence seizures may vary according to individual susceptibility [
Focal and generalized EEG paroxysms in childhood absence epilepsy: topographic associations and distinctive behaviors during the first cycle of non-REM sleep.
]. An experimental study showed bilateral synchronization of the hippocampi during GSWDs in a pharmacologically induced animal model of typical absence seizures [
]. Another study revealed that glucose utilization rates were increased not only in the cortico-thalamic–cortical circuit but also in the limbic area in a genetic animal model of typical absence seizures [
]. These observations demonstrate the involvement of the hippocampus and the interplay between the hippocampal and cortico-thalamo–cortical circuits in typical absence seizures [
Focal and generalized EEG paroxysms in childhood absence epilepsy: topographic associations and distinctive behaviors during the first cycle of non-REM sleep.
Focal and generalized EEG paroxysms in childhood absence epilepsy: topographic associations and distinctive behaviors during the first cycle of non-REM sleep.
], which suggests that the temporal area may be a part of the epileptic network in CAE. Focal temporal electroclinical transformation from typical absence seizures was observed in video-EEG monitoring of two CAE patients [
]. The authors suggested that the temporal lobe may have a function in CAE as a default component of the neuronal network involved with typical absence seizures [
]. Another animal study with a pharmacologically induced model of atypical absence seizures showed that dysplastic changes in the hippocampus were responsible for drug refractoriness [
]. Thus, the circuitry involving the hippocampus and the temporal lobe seems to be associated with drug resistance and pathophysiology in atypical absence seizures. Although recent evidence has shown that temporal lobe involvement is also associated with CAE [
Focal and generalized EEG paroxysms in childhood absence epilepsy: topographic associations and distinctive behaviors during the first cycle of non-REM sleep.
], the clinical implications of this observation for drug responsiveness have not been determined. Based on the role of the temporal lobe in the drug resistance of atypical absence seizures [
Proposal for revised classification of epilepsies and epileptic syndromes. Commission on classification and terminology of the international league against epilepsy.
]. Because of the relative laxity of the ILAE criteria, various types of generalized epilepsy syndromes with frequent absence seizures could be regarded as CAE [
], their inaccurate categorization as CAE may lead to misrepresentation of the prognosis for CAE in clinical studies. Several studies using ILAE criteria reported the remission rate of CAE to be 50–60% [
]. To overcome these issues, we used stricter criteria than the ILAE ones to diagnose CAE in this study. The stricter criteria were proposed by Panayiotopoulos and reviewed by Loiseau et al. [
]. Grosso et al. applied the stricter criteria and revealed a higher remission rate of CAE (82%) compared with studies that applied the ILAE criteria [
]. We also applied the stricter criteria for diagnosis of CAE in this study and observed results similar to Grosso's study, i.e., 88% of the CAE patients enrolled in our study were seizure-free for over 1 year. The concordance between Grosso's study [
] and our study indicate that the stricter criteria may be helpful for the homogenous selection of typical CAE patients. This homogenous selection may add confidence to the results of our study.
This study may be limited because of the clinical data and the analysis method utilized. The sample size in the present study was too small to allow us to draw final conclusions. Additionally, data collection for drug responsiveness was not performed prospectively. Thus, these measures may be not perfect, compared with prospective seizure-free documentation via ambulatory EEG. The selection of differing initial AEDs in each patient may represent another limitation. As initial monotherapy, the superior effectiveness of ethosuximide and valproic acid, compared with lamotrigine, in controlling seizures, was noted as class I evidence [
]. However, there was no difference between the two groups in the types of AEDs administered in this study. Although eLORETA has been validated in the analysis of epilepsy-related data [
]. Additionally, we analyzed only one peak point of an averaged spike. For this reason, our results did not show the entire network producing GSWDs, but just showed the CSD representing the peak point of the spike. Future prospective studies, with a large number of CAE patients that investigate the association between entire networks of GSWDs and drug response may be helpful to clarify our results.
5. Conclusions
Our study investigated, for the first time, whether a hyperexcitable temporal cortex is associated with responsiveness to initial AED treatment in CAE. The present study suggested that a hyperexcitable temporal cortex may be an indicator of poor responsiveness to initial AED treatment. If these results are replicated in larger prospective studies, our results may be helpful in clinical practice by estimating the efficacy of initial AED treatment in naïve CAE patients in advance.
Conflict of interest
None of the authors has any conflict of interest to disclosure.
References
Hirsch E.
Pierre T.
Panayiotopoulos C.P.
Typical absence seizures.
in: Engel J.J. Pedley T.A. Epilepsy: a comprehensive textbook. 2nd ed. Lippincott-Raven Publishers,
Philadelphia1997: 2397-2407
in: Niedermeyer E. Lopes ds Silva F. Electroencephalography: basic principles, clinical applications, and related fields. 6th ed. Wolters Kluwer/Lippincott Williams & Wilkins Health,
Philadelphia2011: 479-540
Physiopathogenesis of generalized epilepsies of organic nature (stereoencephalographic study).
in: Gastaut H. Jasper H.H. Bancaud J. Waltregny A. The physiopathogenesis of the epilepsies. 3rd ed. Charles C Thoma Publisher,
Springfield1969: 158-185
Changes in current-source density of interictal spikes in benign epilepsy of childhood with centrotemporal spikes following treatment with oxcarbazepine.
Focal and generalized EEG paroxysms in childhood absence epilepsy: topographic associations and distinctive behaviors during the first cycle of non-REM sleep.
Proposal for revised classification of epilepsies and epileptic syndromes. Commission on classification and terminology of the international league against epilepsy.
00179-X&id=gr1.jpg){kind=link}
00179-X&id=gr2.jpg){kind=link}