Clinical Neurophysiology 115 (2004) 1169–1177 www.elsevier.com/locate/clinph
Ictal heart rate increase precedes EEG discharge in drug-resistant mesial temporal lobe seizures Giancarlo Di Gennaroa, Pier Paolo Quaratoa, Fabio Sebastianoa, Vincenzo Espositoa,c, Paolo Onoratid,e,*, Liliana G. Grammaldoa, Giulio N. Meldolesia, Addolorata Masciaa, Carolina Falcoa, Ciriaco Scoppettad, Fabrizio Eusebid, Mario Manfredia,b, Giampaolo Cantorea a
Epilepsy Surgery Unit, IRCCS ‘ NEUROMED’ , Pozzilli (IS), Italy Department of Neurological Sciences, University ‘ La Sapienza’ , Rome, Italy c Department of Neurosurgery, University ‘ La Sapienza’ , Rome, Italy d Department of Human Physiology and Pharmacology, University ‘ La Sapienza’ , P.le Aldo Moro 5, 00185 Rome, Italy e Child Developmental Center, San Raffaele Pisana-Tosinvest Sanita`, Rome, Italy b
Accepted 15 December 2003
Abstract Objective: Heart rate (HR) changes, mainly tachycardia, are often observed during seizures originating from the temporal lobe. The aim of this study was to analyze the role of ictal HR changes in localizing both mesial and lateral temporal lobe epilepsy (TLE) in a group of 68 patients. The influence of the gender and the side of epilepsy on HR modulation was also evaluated. Methods: Ictal HR was recorded during prolonged Video-EEG monitoring performed in 68 patients affected by drug-resistant TLE during a non-invasive pre-surgical protocol. According to the electro-clinical correlation, obtained by video-EEG monitoring, one hundred-thirteen seizures (n ¼ 113) and one hundred-forty-four auras (n ¼ 144) were identified and included in the study. Furthermore, the electro-clinical correlation allowed the classification of all the epileptic events (seizures and auras) as having mesial or lateral origin, based on the temporal lobe seizure onset zone. Ictal HR was calculated with respect to the R-R waves, and assessed from 15 sec (s) before (T2 15) to 15 s after (Tþ15) the time of EEG seizure onset (T0). Results: We observed a high incidence (92%) of ictal HR increase in TLE seizures. When the ictal EEG indicated a seizure onset from the mesial temporal structures, the onset of ictal HR increase preceded by about 5 s the EEG ictal onset (SD ^ 18.4), whereas the onset of HR increase coincided with the onset of EEG discharges (SD ^ 14.8) when the ictal EEG indicated the onset of seizures from the lateral temporal structures. No significant differences were found between male and female patients; and between right and left TLE. Conclusions: Our findings show that ictal HR increase, preceding the onset of the EEG discharge, is associated with ictal EEG seizure pattern defining temporal lobe seizures originating from the mesial temporal lobe structures; this association suggests that the HR changes may be coupled to the functional impairment of neural circuits involved in sympathetic cardiovascular regulation, in the mesial temporal lobe structures. Further studies investigating the relationship between intracranial EEG monitoring and ECG recording are worthwhile, to confirm our results and to give further indications on the pathogenesis of ictal HR abnormalities. q 2004 International Federation of Clinical Neurophysiology. Published by Elsevier Ireland Ltd. All rights reserved. Keywords: Heart rate; Temporal lobe epilepsy; Presurgical evaluation; Video EEG; Epilepsy surgery
1. Introduction The ultimate goal of the non-invasive pre-surgical evaluation is to localize the epileptogenic zone (Rosenow and Luders, 2001). Long-term non invasive Video-EEG * Corresponding author. Tel.: þ39-06-4991-0896; fax: þ 39-06-49910851. E-mail address:
[email protected] (P. Onorati).
recording has a crucial role in the identification of the epileptogenic zone, since it provides both EEG interictal data and ictal electro-clinical correlations (Gates, 1986; Bancaud and Talairach, 1991; Sperling et al., 1992; Wieser and Williamson, 1993; Kaplan and Lesser, 1996; Clemens and Menes, 2000; Holmes et al., 2000; Mintzer et al., 2002). Clinical, neuroradiological and neuropsychological evaluations, in combination with video-EEG monitoring, are successfully used to localize the epileptogenic zone
1388-2457/$30.00 q 2004 International Federation of Clinical Neurophysiology. Published by Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.clinph.2003.12.016
1170
G. Di Gennaro et al. / Clinical Neurophysiology 115 (2004) 1169–1177
in the majority of patients with temporal lobe epilepsy (TLE) without performing intracranial investigations (Shorvon, 2000). A diagnostic grid, which proposes the use of clinical and para-clinical criteria to identify the epileptogenic zone, has been recently developed in our Epilepsy Surgery Unit (ESU) (Di Gennaro et al., 2003). The main advantage by using of this grid is that 3 different subtypes of TLE (i.e. mesial, lateral, mesiolateral) can be identified, and the surgery can be planned on the basis of the extension of the epileptogenic zone (Di Gennaro et al., 2003). Although the ictal EEG patterns can be used to distinguish between the subtypes of TLE (Ebersole and Pacia, 1996; Pacia and Ebersole, 1997; Foldvary et al., 2001), the ictal clinical signs alone do not provide any convincing information about the epileptogenic zone. One of these ictal clinical signs is the modification of the heart rate (HR), which often occurs during a temporal lobe seizure (Marshall et al., 1983; Blumhardt et al., 1986; Garcia et al., 2001). Previous works refer a high incidence of tachycardia and, more rarely, bradycardia, in TLE (Marshall et al., 1983; Blumhardt et al., 1986; Garcia et al., 2001; Leutmezer et al., 2003), but the localizing significance and timing of their appearance in the development of temporal lobe seizures have not been fully investigated (Nashef et al., 1996; Garcia et al., 2001; Ansakorpi et al., 2002; Kirchner et al., 2002; Leutmezer et al., 2003). Moreover, the relationship between ictal HR changes and specific temporal lobe structures remains unclear (Blumhardt et al., 1986; Garcia et al., 2001). Therefore, the aims of this study were to investigate both the type and the onset of ictal HR changes and to evaluate whether the HR modifications were correlated with different seizure onset zones (Rosenow and Luders, 2001), namely mesial or lateral from 68 patients with drug resistant TLE scheduled for surgery. We also evaluated whether there was a link between HR changes, side of the epilepsy and the patients’ gender.
2.2. Pre-surgical diagnostic protocol All patients underwent the non invasive pre-surgical protocol previously described in detail (Di Gennaro et al., 2003). In brief, the protocol included: (1) continuous long-term intensive, diurnal and nocturnal, Video-EEG monitoring (Telefactor Corp., Conshohoken, PA); (2) neuropsychological evaluation; (3) psychiatric evaluation; (4) computed tomography scan and morphologic magnetic resonance imaging (MRI); (5) multimodal evoked potential recording, transcranial magnetic stimulation (Rossini et al., 2002) and visual field examination (when required, two patients); (6) mono- or bilateral Wada test (when required, 6 patients). All patients had cardiac evaluation in order to exclude any cardiac disorders that could impair the analysis of ictal HR changes. No patients had history or evidence of abnormalities of the cardiovascular system or any other disorder that could affect the autonomic nervous system. The Video-EEG recording technique was in accordance with international guidelines developed by the American Electroencephalographic Society for Video-EEG monitoring of epileptic patients (American Electroencephalography Society, 1986). Scalp EEG analysis was done using bipolar longitudinal-transverse and referential montages: EEG traces were displayed on a high definition monitor (2024 £ 1860 d/i), and two filter settings, i.e. LFF 1.3 and HFF 50 Hz, were used. The EEG recording of each patient was examined in order to evaluate the presence of background abnormalities, both slow and epileptiform interictal activity, ictal discharges and post-ictal slowing. The onset of the ictal discharge was defined as the first ictal EEG change which lasted at least 3 sec (s) and was localized over, or lateralized to, the side of the epileptogenic zone (Vossler et al., 1998). The ictal discharges were firstly divided into right and left temporal lobe seizures; those ictal discharges which spread from the hemisphere of origin during the first 30 s from the onset were excluded from the analysis. All EEG recordings were analyzed by two examiners (G.D.G, and P.P.Q.) who were kept blind to each other’s findings.
2. Patients and methods 2.3. Seizure classification 2.1. Patients From a series of 166 consecutive patients affected by medically refractory partial epilepsy and referred to us for presurgical assessment, 83 patients affected by TLE were identified. Out of all these TLE patients, 68 were included in the study. There were 26 females and 42 males. Their mean age was 33.2 (SD ^ 19) years while the mean duration of their epilepsy was 18.3 (SD ^ 9.7, range 1 –46) years. All patients were right-handed and had taken more than two antiepileptic drugs. They were referred to the ESU of IRCCS NEUROMED, Pozzilli (IS), Italy, between September 1998 and November 2002.
The electro-clinical semiology of the all video-EEG recorded epileptic events was analyzed in order to define the seizure onset zone. To achieve this goal, the combination ictal EEG features (scalp regions of onset, frequency and propagation patterns of discharge) and the synchronized assessment of the patient’s behavior during seizures was strictly taken in account; as for the final judgment regarding the seizure classification, clinical and EEG ictal features always agreed in suggesting the seizure onset zone. In this combined frame, the EEG frequency pattern was stressed as objective criterion for seizure classification, according to Garcia et al. (2001). Therefore, the ictal discharges of each
G. Di Gennaro et al. / Clinical Neurophysiology 115 (2004) 1169–1177
side were divided into two groups: group 1 (5 – 9 Hz), which suggests the seizure onset zone in the mesial aspects of the temporal lobe, and group 2 (2 – 4 Hz), which suggests the seizure onset zone in the lateral aspects of the temporal lobe (Ebersole and Pacia, 1996; Pacia and Ebersole, 1997; Foldvary et al., 2001). Only those seizures in which the ictal EEG pattern and the ictal clinical features were not conflicting in suggesting an involvement of mesial (i.e. clinical symptoms/signs such as psychic-vegetative manifestation, early oro-alimentary automatisms and arm dystonia contralateral to the epileptogenesis, and 5 – 9 Hz EEG discharge) or lateral (i.e. clinical symptoms/signs such as auditory or vestibular manifestations, deja vu-deja` vecu, and 2– 4 Hz EEG discharge) aspects of the temporal lobe were classified in the two groups and considered for the study. On the basis of this, 15 out of the 83 TLE patients were excluded from the study, including those patients whose seizures did not come clearly within the definition of either group (disagreement between EEG and clinical findings, EEG pattern characterized by mixed frequencies or not interpretable) and those patients who fell simultaneously into both groups (patients with both electro-clinical mesial and lateral seizures). Auras (simple partial epileptic episodes characterized only by the initial sensation of a typical seizure in a given patient, with no overt signs and that the patient is aware of and able to recollect them) in which an ictal change was detectable on EEG were also included in the study. When in a given patient auras were not associated to a ‘ build-up’ discharges but were characterized by a focal flattening on EEG (as frequently occurs), they were placed in the group (either group 1 or 2) as identified by the electro-clinical correlation obtained in the case of overt seizures accompanied by an EEG ‘ build-up’ pattern recorded in that patient. In instances of disagreements between the examiners, a final consensus was achieved after discussion. Finally, the number of female and male patients were counted in each group. 2.4. Patient classification Using the information obtained from the pre-surgical protocol, 35 patients out of the 68 included in the study (51.5%; 18 male and 17 female) were diagnosed with right TLE and the remaining 33 (48.5%; 24 male and 9 female) with left TLE. According to our diagnostic grid which is based on the correlation between anatomic and electroclinical data to identify the different epileptogenic zones within the TLE and consequently to plan the surgical strategy (Di Gennaro et al., 2003), the sample of 68 patients selected for this study were further classified as mesiolateral TLE (n ¼ 51), mesial TLE (n ¼ 11) and lateral TLE (n ¼ 6) patients. Surgery was performed in 60/68 TLE patients. Histological examination performed after surgery revealed
1171
32 mesial temporal sclerosis, 4 focal cortical atrophy, 2 dysembryoplastic neuroepithelial tumors, 7 dysplasias, 10 low-grade gliomas and 3 cavernomas. In two patients no abnormality was found. Out of the 44 patients who completed a 1 year follow-up, 39 patients were seizurefree (Engel class IA), 4 exhibited rare and non-disabling simple partial episodes (Engel class IB), and one patient continued to suffer from seizures and did not have any improvement in quality of life (Engel class IV) (Engel, 1987). Out of the remaining 16 patients who were assessed before the completion of 1 year of follow-up, 15 patients were reported as seizure-free and one patient as having rare auras. 2.5. Analysis of electrocardiogram (ECG) during seizures The ECG was recorded in all patients using two supraclavicular leads (left and right) and was reviewed in bipolar montage. The ECG was reported by two examiners (F.S. and C.F.) who were blind to the seizure EEG pattern. For each seizure included in the study the following data were considered: (1) clinical onset, (2) EEG onset (T0), and (3) EEG pattern. A baseline (B) HR was then calculated using the mean of the HR during the first 15 s of the 60 s immediately preceding T0. The ictal HR was assessed from 15 s before (T2 15) to 15 s after (Tþ15) with respect to T0, using the R-R wave intervals. Both the B and ictal HR were expressed as beats/min and were calculated in each seizure. For each seizure evaluated we performed a mathematical analysis of the ECG R-R wave intervals from 15 s before (T2 15), to 15 s after (Tþ15) with respect to T0, by using an interpolating second order polynomial. For each curve generated by the polynomial we analyzed the sign of firstderivation calculated for each R-R intervals. Transition from a horizontal asymptote to an oblique one, positive or negative, indicated HR increase or HR decrease, respectively. We defined as onset of ictal HR change the point, in the T2 15 –Tþ15 epoch, from which an asymptotic transition maintains the same sign until the end of this epoch. We excluded from the study ictal ECGs where the corresponding EEG could not be interpreted because of artifacts. All EEG data from 60 s before to 60 s after T0 were transformed into an EDF format (European Data Format file) and the analysis was performed using MATLAB (version 6.1, The Mathworks, Natick, MA). The ECG signal was processed automatically to detect the R-R interval and to generate the HR variations curve, which was created by plotting the HR at 5 s intervals. A MATLAB routine was implemented in order to analyze the ECG and to generate a matrix of data. In this matrix, the ith column represents the generic temporal step, and the jth row the values which were obtained by calculating the difference between the value represented at
1172
G. Di Gennaro et al. / Clinical Neurophysiology 115 (2004) 1169–1177
an ith step and the preceding step (i 2 1). This matrix indicates the punctual slope variations for each analyzed curve (delta analysis). 2.6. Statistical analysis Two statistical procedures were applied. (i) Inter-group analysis: we used four-way ANOVA to look at the differences between group 1 and group 2, respectively indicating mesial and lateral TLE seizures, female and male patients, right and left sided seizures at three time points (T0, T2 15, Tþ15). P value , 5 £ 10-2 was considered as significant. The inter-group analysis was applied also to the slope of the mean curves of the HR. (ii) Intra-group analysis: in the mesial and lateral TLE seizures group, the time of the onset of the HR change was calculated with respect to the B HR. The Mauchly’s test of sphericity was used to test the hypothesis that the covariance matrix of the transformed variables has a constant variance on the diagonal and zeros off the diagonal. The correction of the degrees of freedom was made by using the Greenhouse-Geisser procedure; Duncan test was performed for post-hoc comparisons (PD , 5 £ 1022 ).
3. Results A total of 351 epileptic episodes were recorded from the 68 patients studied. Of these episodes only 257 were included in the study and they represented 113 seizures and 144 auras (Fig. 1). In the group of mesiolateral TLE patients, the epileptic episodes were either mesial (95) or lateral (97) while in the groups of mesial TLE and lateral TLE patients all events were respectively mesial (42) and lateral (23). In each of the 257 epileptic episodes, HR variation curves were generated. Ictal HR change was detected as increase in 237 cases (92%). In 10 seizures (4%) we detected a transient HR decrease (lasting 3 – 5 R-R intervals, decreasing from 77.3 to 76.5 beats/min). In 10 seizures (4%)
no ictal HR changes were observed. The 10 cases in which ictal HR decrease was detected were not considered for analysis because of both the brief duration of the ictal HR change and the exiguity of sampled data. In the right TLE the mean HR was 77.5 beats/min at T2 15 and increased to 82.8 beats/min at Tþ15. In left TLE seizures it increased from 75.6 to 79.7 beats/min (Table 1). The HR changes were also divided into those which occurred in the mesial and lateral TLE, and those which occurred in the male and female group (Table 1). The inter-group analysis showed that in the male group there was a significantly greater increase in the HR at the right mesial temporal lobe seizures when compared with the left-sided mesial temporal lobe seizures 10 s after T0. Moreover, there were no significant differences between all the other groups examined (Table 2). The intra-group analysis showed that the onset of ictal HR increase occurred earlier in the mesial than in the lateral temporal seizures. In fact, in 98% out of the total mesial seizures included in the study, ictal HR increase occurred 5 s before the ictal EEG discharge. In contrast, in 95% of the total lateral seizures, ictal HR increase coincided with the onset of the ictal EEG discharge (Table 2). The slope analysis (delta analysis) of the curves indicating ictal HR variations in both mesial and lateral seizure groups showed the onset of ictal HR increase respectively from 5 s before T0 in the mesial group and at T0 in the lateral group (Fig. 4). 3.1. Representative patients 3.1.1. Patient 1 A.F., a 43-year-old man, had seizure onset at age 14 years. He was born after an unremarkable pregnancy and delivery and he had prolonged febrile seizures at age 2 years. He experienced weekly seizures consisting of a rising epigastric sensation, fear and tachycardia, followed by loss of consciousness, oroalimentary automatisms and dystonia of the left hand. MRI revealed a right-sided mesial temporal sclerosis. Video-EEG monitoring showed interictal spikes on the right anterior temporal regions. Ictal EEG showed
Fig. 1. Classification of the temporal lobe epileptic episodes included in the study.
G. Di Gennaro et al. / Clinical Neurophysiology 115 (2004) 1169–1177
1173
Table 1 Baseline rate and ictal heart rate changes analysis within each different seizure subgroups Seizure subgroups
Right seizures Left seizures Mesial seizures Lateral seizures Right seizures in males Left seizures in males Right seizures in females Left seizures in females
Baseline (B) (beats/min) Mean (SD)
T2 15 (beats/min)
T0 (beats/min)
Tþ15 (beats/min)
Statistical analysis
Mean (SD)
Mean (SD)
Mean (SD)
(B 2 Tþ15)
76.9 (16.5) 74.8 (13.9) 74.6 (16.9) 77.3 (13.3) 77.6 (15.8) 74.0 (15.1) 76.4 (17.1) 76.8 (10.3)
77.5 (16.7) 75.6 (14.1) 75.5 (17.1) 77.7 (13.4) 77.4 (15.5) 75.5 (14.9) 77.6 (17.5) 75.4 (11.8)
78.8 (18.7) 76.7 (15.1) 78.7 (18.8) 77.8 (14.8) 80.3 (17.1) 75.9 (15.4) 77.9 (19.7) 78.7 (14.1)
82.8 (19.8) 79.7 (15.3) 80.5 (19.8) 82.1 (15.2) 83.5 (17.9) 78.6 (15.3) 82.3 (21) 82.2 (15.2)
Fð6; 8Þ 5 22:6; MSE 5 20:4; P < 1 3 1025 ; PD < 4 3 1026 Fð6; 7Þ 5 9:4; MSE 5 29:6; P < 1 3 1025 ; PD < 4 3 1026 Fð7; 9Þ 5 17:7; MSE 5 29:4; P < 1 3 1025 ; PD < 5 3 1026 Fð7; 8Þ 5 10:7; MSE 5 29; P < 1 3 1025 ; PD < 5 3 1026 Fð7; 4Þ 5 9:2; MSE 5 26; P < 1 3 1025 ; PD < 4 3 1026 Fð7; 6Þ 5 5:7; MSE 5 29:8; P < 1 3 1025 ; PD < 5 3 1026 Fð7; 6Þ 5 12:3; MSE 5 23:1; P < 1 3 1025 ; PD < 5 3 1026 Fð7; 2Þ 5 4:7; MSE 5 46:2; P < 1 3 1024 ; PD < 2 3 1023
Statistical significance is indicated by boldface type.
flattening on anterior right temporal regions followed by rhythmic 6– 7 Hz, ‘ mesial’ temporal discharges. The ictal HR showed an increase 5 s before the onset of the ictal EEG discharge (Fig. 2). 3.1.2. Patient 2 S.S., a 33-year-old man, had seizure onset at age 28 years. He experienced weekly seizures consisting of deja` vecu followed by loss of consciousness, motionless stare with no automatisms, and postictal dysphasia. MRI revealed an occupying space lesion (ganglioglioma) in neocortical structures of the left temporal lobe. Video-EEG monitoring showed interictal sharp waves on the left temporal regions. Ictal EEG showed flattening on left temporal regions followed by rhythmic 3 –4 Hz ‘ lateral’ temporal discharges. The ictal HR showed an increase ‘ coinciding’ with the onset of the ictal EEG discharge (Fig. 3).
4. Discussion The overall objective of this study was to explore the role of the ictal HR changes in localizing both mesial and lateral TLE. Taken together, our findings indicate a high incidence (96%) of ictal HR changes in TLE seizures, these changes being mostly characterized by an ictal increment in the HR. Moreover, this HR ictal increment significantly precedes the ictal EEG seizure pattern underlying temporal lobe seizures originating from the mesial aspect of the temporal lobe. Two main considerations are needed in order to fully address the interpretation of these findings. In the present experimental context, isolated auras were pooled together with the complex partial seizures recorded in the same patient. This was firstly, because it is largely accepted in literature that the initial seizure semiology usually provides valuable information about the seizure onset zone
Table 2 Ictal heart rate changes differences between different seizure subgroups and inter-group analysis Statistical analysis Inter-group analysis: seizure subgroups matched Right vs. left seizures Mesial vs. lateral seizures Right vs. left seizures, males vs. females Right mesial vs. left mesial seizures Mesial seizures in males vs. mesial seizures in females Lateral seizures in males vs. lateral seizures in females Right lateral seizures vs. left lateral seizures Right mesial seizures in males vs. left mesial seizures in males Right mesial seizures in females vs. left mesial seizures in females Intra-group analysis Mesial seizures
Lateral seizures
Fð7; 2Þ ¼ 0:5; MSE ¼ 29:2; P , 8:3 £ 1021 Fð7; 2Þ ¼ 0:4; MSE ¼ 29:2; P , 8:7 £ 1021 Fð21; 2Þ ¼ 1:1; MSE ¼ 29:2; P , 3:9 £ 1021 Fð7; 9Þ ¼ 1:2; MSE ¼ 29:4; P , 3:1 £ 1021 Fð7; 9Þ ¼ 0:5; MSE ¼ 29:6; P , 8:3 £ 1021 Fð7; 8Þ ¼ 0:2; MSE ¼ 29:2; P , 9:8 £ 1021 Fð7; 8Þ ¼ 0:3; MSE ¼ 29:2; P , 9:3 £ 1021 Fð21; 9Þ 5 2; MSE 5 28:8; P < 5 3 1023 ; PD < 2:8 3 1023 (10 s after T0) Fð7; 4Þ ¼ 0:8; MSE ¼ 29:2; P , 3 £ 1021
Ictal HR increase occurred 5 s (SD 6 18.4) before the onset of the ictal EEG discharge ½Fð7; 9Þ 5 17:7; MSE 5 29:4; P < 1 3 1025 ; PD < 5 3 1026 Ictal HR increase coincided with the onset (SD 6 14.8) of the ictal EEG discharge ½Fð7; 8Þ 5 10:7; MSE 5 29; P < 1 3 1025 ; PD < 5 3 1026
Right mesial seizures showed significantly greater increase in HR when compared with the left-sided mesial seizures 10 s after T0, only in males. In mesial seizures ictal HR increase occurred 5 s before the onset of the ictal EEG discharge while in lateral seizures ictal HR increase coincided with the onset of the ictal EEG discharge. Statistical significance is indicated by boldface type.
1174
G. Di Gennaro et al. / Clinical Neurophysiology 115 (2004) 1169–1177
Fig. 2. Ictal EEG of right temporal seizure with ‘ mesial’ discharge pattern. The ictal HR increase is visible 5 s before the ictal EEG onset. Number ‘ 1’ : ictal HR increase onset; number ‘ 2’ : ictal EEG onset.
(Luders and Awad, 1992; Palmini and Gloor, 1992; Henkel et al., 2002) as it may be assumed that in these cases the ictal discharge involves a limited area of the cerebral cortex. In our study the clinical semiology of the isolated auras strictly corresponded to those which preceded seizures with overt signs. Secondly, although a scalp EEG ictal ‘ build-up’ pattern may be highly associated with a particular seizure onset zone (i.e. hippocampus), it is not a direct manifestation of hippocampal activity and it does become visible only when adjacent temporal cortex is involved in the seizure and thus, relatively late with respect to the seizure onset (Pacia and Ebersole, 1997). In this view, the focal electrodecremental pattern which almost always represents the only ictal EEG finding underlying isolated auras as well as the initial EEG feature of the complex partial seizures, may be considered the scalp counterpart of a discharge strictly confined to ictal onset cortical areas. Therefore, the localizing value of an aura may be established only analyzing the evolution of an aura into other seizure types. The second consideration concerns the apparent discrepancy between the classification system of seizures
(mesial or lateral) and of the TLE epileptogenic zone (mesial, lateral or mesiolateral) that rises from intrinsic conceptual differences proper of the ‘ objects’ to be classified (i.e. seizure onset zone and epileptogenic zone). Indeed, the epileptogenic zone, which may be considered the area of the cortex indispensable for the generation of seizures and that it should be removed for the seizure disappearance, not infrequently is more extensive than the seizure onset zone (Rosenow and Luders, 2001). Accordingly, many patients belonging to the studied population were classified having mesiolateral TLE and consequently underwent extensive temporal lobectomy, despite of the occurring seizures pointing to a more limited seizure onset zone (mesial or lateral). Our finding confirms previous reports which have found a high incidence of HR abnormalities, especially tachycardia, in TLE (Marshall et al., 1983; Blumhardt et al., 1986; Smith et al., 1989; Schernthaner et al., 1999; Garcia et al., 2001; Tinuper et al., 2001; Leutmezer et al., 2003). However, comparisons between studies have to be treated with caution, since the methodological approaches applied
G. Di Gennaro et al. / Clinical Neurophysiology 115 (2004) 1169–1177
1175
Fig. 3. Ictal EEG of left temporal seizure with ‘ lateral’ discharge pattern. The ictal HR increase coincides with the ictal EEG onset. Number ‘ 1’ : ictal HR increase onset; number ‘ 2’ : ictal EEG onset.
and the characteristics of the cohorts were different. For example, some authors (Li et al., 1995; Galimberti et al., 1996) report ictal tachycardia in just over a third of the TLE patients using in-patient video-EEG recording, while some others, using the same method, report ictal tachycardia in almost 100% of temporal lobe seizures (Marshall et al., 1983). Finally, Blumhardt et al. (1986) report a high incidence of ictal tachycardia in TLE (90% of the patients) by simultaneous monitoring of ambulatory cassette EEG and ECG. Moreover, since ictal bradycardia seems to occur mainly in patients who are at least 60 years old (Tinuper et al., 2001), the relatively low mean age of our cohort could explain the infrequent cases of bradycardia in our study. Although the occurrence of tachycardia in TLE is well known, the relationship between the onset and the involvement of temporal lobe structures has not been widely investigated (Blumhardt et al., 1986; Garcia et al.,
Fig. 4. Slope analysis (delta analysis). The curves indicating ictal HR variations in both mesial and lateral seizure groups show the onset of ictal HR increase, respectively from 5 s before T0 in mesial group and at T0 in lateral group.
1176
G. Di Gennaro et al. / Clinical Neurophysiology 115 (2004) 1169–1177
2001). We found that in most seizures in which the ictal EEG indicated an onset in the mesial temporal structures, the onset of HR increase occurred significantly earlier than in the lateral temporal lobe seizures. In particular, in the mesial seizures the ictal HR increase was clearly detectable before both EEG and clinical ictal onset, while in almost all the lateral temporal lobe seizures the start of HR increase coincided with the EEG ictal onset. Previous reports concerning autonomic symptoms during epileptic seizures (Schernthaner et al., 1999; Baumgartner et al., 2001) show that sinus tachycardia preceded EEG seizure onset for an average of 18.7 s in 76.1% of all focal seizures on surface EEG, and for an average of 11.0 s in 45.7% of seizures on invasive subdural EEG; early tachycardia occurs significantly more often in temporal than in frontal lobe origin seizures. These results were confirmed in a study on patients affected by partial epilepsy who underwent video-EEG monitoring (Leutmezer et al., 2003); the authors concluded that the incidence as well as the amount of ictal HR increase was significantly more pronounced in patients with mesial TLE as compared with those with non-lesional TLE or extratemporal epilepsy. In these reports, however, a distinction between mesial or lateral temporal seizures based on seizure EEG pattern was not made and therefore a different behavior of the ictal HR change depending on different seizure onset zone in temporal lobe cannot be inferred. Our findings confirmed and extended the results of the previous studies which have reported the onset of ictal tachycardia preceding the onset of EEG discharge. Blumhardt et al. (1986) suggested a relationship between tachycardia and seizure onset in the deep limbic temporal structures, with a possible role for HR changes in the diagnosis of TLE. The difference with our analysis was the investigation of the onset of ictal HR change in both mesial and lateral temporal lobe seizures. We found that only in the mesial seizures, ictal HR change preceded the EEG ictal discharge. This result supports the hypothesis, which was proposed by a preliminary study in 28 patients (Garcia et al., 2001), of a strong relationship between earlier ictal tachycardia and mesial temporal lobe seizures. We propose that this relationship may reflect the functional impairment of neural circuits involved in sympathetic cardiovascular regulation in the mesial temporal structures. On the other hand, the relatively late appearance of ictal HR increase in the course of lateral temporal seizures may infer that the ictal discharge starts in the lateral structures and successively spreads to the mesial structures. These results are in agreement with a previous study (Van Buren and Ajmone-Marsan, 1960) that emphasized that autonomic features of TLE occur early in the course of the seizures and precede the other clinical manifestations. Moreover, several experimental (Applegate et al., 1983; Gelsema et al., 1993) and clinical studies (Kahane et al., 1999) suggested that the mesial temporal structures and, in
particular, the insulo-mesiotemporal-fronto-orbital neural system with its connections to the subcortical areas, play a role in the genesis of ictal HR changes. We also investigated the differences in HR changes between groups, and found that in the males there was a significantly greater increase in HR in the mesial right-sided seizures than in the left-sided that occurred about 10 s after the EEG ictal onset. However, this observation needs to be confirmed by further investigations. This is because it occurs 10 s after the EEG ictal onset and may be related to confounding factors, such as ictal heart changes caused by movement or emotional state, bilateralization of the discharge, etc. A recent study (Kirchner et al., 2002) analyzed 21 TLE patients and showed that males affected by right TLE had a significantly greater increase in HR in the early part of the seizure than males with left TLE and females with TLE. This suggested gender differences and brain asymmetry in HR control. However, in contrast to our study, a classification of TLE seizures into mesial and lateral was not performed. Unlike Kircher’s report, we failed to find gender differences. The following factors may explain these differences: firstly, we evaluated a greater number of seizures; secondly, we considered the seizure onset to be the first ictal EEG change and not the first ictal clinical sign. Nevertheless, several experimental and clinical reports suggested that the cardiovascular modulation is mainly based in the right hemisphere (Oppenheimer et al., 1992; Yoon et al., 1997; Hilz et al., 2001). Therefore, further studies investigating the relationship between intracranial EEG monitoring (Kahane et al., 1999) with simultaneous scalp EEG and ECG recording are needed to clarify the pathogenesis of ictal HR abnormalities.
Acknowledgements We are deeply grateful to Dr Olga Ciccarelli, NMR Research Unit, Institute of Neurology, Queen Sqare, London, UK, for helpful suggestions and for English revision. We also thank Drs Francesco Mari and Teresa Giampa` for the helpful management of the patients and the team of neurophysiopathology technicians (Antonella Addesso, Lucia Cacciola, Carmen Grappa, Fabio Orabona, Raffaella Ponticelli), for performing meticulous Video-EEG recordings in all patients included in the study.
References American Electroencephalography Society, Guidelines for long-term monitoring for epilepsy. J Clin Neurophysiol 1986;63:107 –11. Ansakorpi H, Korpelainen JT, Huikuri HV, Tolonen U, Myllyla VV, Isojarvi JIT. Heart rate dynamics in refractory and well controlled temporal lobe epilepsy. J Neurol Neurosurg Psychiatry 2002;72:26 –30. Applegate CD, Kapp BS, Underwood MD, McNail CL. Autonomic and somatomotor effects of amygdala central stimulation in awake rabbits. Phys Behav 1983;31:353–60.
G. Di Gennaro et al. / Clinical Neurophysiology 115 (2004) 1169–1177 Bancaud J, Talairach J. Se´me´iologie clinique des crises du lobe temporal. Crises epileptiques et epilepsies du lobe temporal, tome II. Gentilly: Documentation Medicale Labaz; 1991. Baumgartner C, Lurger S, Leutmezer F. Autonomic symptoms during epileptic seizures. Epileptic Disord 2001;3:103–16. Blumhardt L, Smith P, Owen L. Electrocardiographic accompaniments of temporal lobe epileptic seizures. Lancet 1986;10:1051 –6. Clemens B, Menes A. Sleep spindle asymmetry in epileptic patients. Clin Neurophysiol 2000;111:2155–9. Di Gennaro G, Quarato PP, Onorati P, Colazza GB, Mari F, Grammaldo L, Ciccarelli O, Meldolesi NB, Sebastiano F, Manfredi M, Esposito V. Localizing significance of temporal intermittent rhythmic delta activity (TIRDA) in drug-resistant focal epilepsy. Clin Neurophysiol 2003;114: 70–8. Ebersole JS, Pacia SV. Localization of temporal lobe foci by ictal EEG patterns. Epilepsia 1996;37:386 –99. Engel Jr J. Outcome with respect to epileptic seizures. In: Engel J Jr, editor. Surgical treatment of the epilepsies. New York: Raven; 1987. p. 553–71. Foldvary N, Klem N, Hammel J, Bingaman W, Najm I, Luders H. The localizing value of ictal EEG in focal epilepsy. Neurology 2001;57: 2022–8. Galimberti C, Marchioni E, Barsizza F, Manni R, Sartori I, Tartara A. Partial epileptic seizures of different origin variably affect cardiac rhythm. Epilepsia 1996;37:742–7. Garcia M, D’Giano C, Estelles S, Leiguarda R, Rabinowicz A. Ictal tachycardia: its discriminating potential between temporal and extratemporal seizure foci. Seizure 2001;10:415–9. Gates RJ. Epilepsy pre-surgical evaluation in era of intensive neurodiagnostic monitoring. In: Gummit RJ, editor. Intensive neurodiagnostic monitoring. Advances in neurology, vol. 46. New York: Raven; 1986. p. 227–48. Gelsema AJ, Copeland NE, Drolet G, Bachelard H. Cardiovascular effects of neuronal activation of the extended amygdala in rats. Brain Res 1993;626:156 –66. Henkel A, Noachtar S, Pfander M, Luders HO. The localizing value of the abdominal aura and its evolution. Neurology 2002;58:271–6. Hilz MJ, Dutsch M, Perrine K, Nelson PK, Rauhut U, Devinsky O. Hemispheric influence on autonomic modulation and baroreflex sensitivity. Ann Neurol 2001;49:575–84. Holmes MD, Kutsy RL, Ojemann GA, Wilensky AJ, Ojemann LM. Interictal, unifocal spikes in refractory extra-temporal epilepsy predict ictal origin and post-surgical outcome. Clin Neurophysiol 2000;111: 1802–8. Kahane P, Di Leo M, Hoffmann D, Munari C. Ictal bradycardia in a patient with a hypothalamic hamartoma: a stereo-EEG study. Epilepsia 1999; 40:522–7. Kaplan PW, Lesser RP. Noninvasive EEG. In: Wyllie E, editor. The treatment of epilepsy. Principles and practice, 2nd. Baltimore: Williams and Wilkins; 1996. p. 976–87. Kirchner A, Pauli E, Hilz MJ, Neundorfer B, Stefan H. Sex differences and lateral asymmetry in heart rate modulation in patients with temporal lobe epilepsy. J Neurol Neurosurg Psychiatry 2002;73:73–5.
1177
Leutmezer F, Schernthaner C, Lurger S, Potzelberger K, Baumgartner C. Electrocardiographic changes at the onset of epileptic seizures. Epilepsia 2003;44:348–54. Li LM, Roche J, Sander JW. Ictal ECG changes in temporal lobe epilepsy. Arquiv Neuro-psiquiatria 1995;53:619–24. Luders HO, Awad IA. Conceptual considerations. In: Luders HO, editor. Epilepsy surgery. New York: Raven; 1992. p. 51–62. Marshall D, Westmoreland B, Sharbrough F. Ictal tachycardia during temporal lobe seizures. Mayo Clin Proc 1983;58:443–6. Mintzer S, Nicholl JS, Stern JM, Engel J. Relative utility of sphenoidal and temporal surface electrodes for localization of ictal onset in temporal lobe epilepsy. Clin Neurophysiol 2002;113:911–6. Nashef L, Walker F, Allen P, Sander JW, Fish DR. Apnea and bradycardia during epileptic seizures: relation to sudden death in epilepsy. J Neurol Neurosurg Psychiatry 1996;60:297– 300. Oppenheimer SM, Gelb A, Girvin JP, Hachinsky VC. Cardiovascular effects of human insular cortex stimulation. Neurology 1992;42: 1727–32. Pacia SV, Ebersole JS. Intracranial EEG substrates of scalp ictal patterns from temporal lobe foci. Epilepsia 1997;38:642–54. Palmini A, Gloor P. The localizing value of auras in partial epilepsies. Neurology 1992;42:801–8. Rosenow F, Luders H. Presurgical evaluation of epilepsy. Brain 2001;124: 1683–700. Rossini PM, Berardelli A, Cantello R. Neuromagnetic methods and transcranial magnetic stimulation for testing sensorimotor cortex excitability. In: Guerrini R, Icardi J, Andermann F, Hallet M, editors. Epilepsy and movement disorders. Cambridge: Cambridge University Press; 2002. p. 59–76. Schernthaner C, Lindinger G, Potzelberger K, Zeiler K, Baumgartner C. Autonomic epilepsy – the influence of epileptic discharges on heart rate and rhythm. Wien Klin Wochenschr 1999;111:392–401. Smith PEM, Howell SJL, Owen L, Blumhardt L. Profile of instant heart rate during partial seizures. Clin Neurophysiol 1989;72:207 –17. Shorvon S. Handbook of epilepsy treatment. Oxford: Blackwell Science; 2000. Sperling MR, O’Connor MJ, Saykin AJ, Phillips MD, Bridgeman PA, French JA, Gonatas N. A non-invasive protocol for anterior temporal lobectomy. Neurology 1992;42:416 –22. Tinuper P, Bisulli F, Cerullo A, Carcangiu R, Marini C, Pierangeli G, Cortelli P. Ictal bradycardia in partial epileptic seizures: autonomic investigation in three cases and literature review. Brain 2001;124:2361–71. Van Buren JM, Ajmone-Marsan G. A correlation of autonomic and EEG components in temporal lobe epilepsy. Arch Neurol 1960;3:683–703. Vossler DG, Kraemer DL, Knowlton RC, Kjos BO, Rostad SW, Wyler AR, Haltiner AM, Hasegawa H, Wilkus RJ. Temporal ictal electroencephalographic frequency correlates with hippocampal atrophy and sclerosis. Ann Neurol 1998;43:756–62. Wieser HG, Williamson PD. Ictal semiology. In: Engel J Jr, editor. Surgical treatment of the epilepsies. New York: Raven; 1993. p. 161 –72. Yoon BW, Morillo CA, Cechetto DF, Hachinsky V. Cerebral hemispheric lateralization in cardiac autonomic control. Arch Neurol 1997;54: 741– 4.
