through the ventral amygdalofugal pathway (Hopkins & Holstege, 1978; Price & Amaral,. 1981; deVito & Smith, 1982) which has been shown to be essential for ...
Pre-Print version – Brain 1994, 117:739-746
Relationship between atrophy of amygdala and ictal fear in temporal lobe epilepsy.
Fernando Cendes, Frederick Andermann, Pierre Gloor, Antonio Gambardella, Iscia Lopes-Cendes, Craig Watson, Alan Evans, Stirling Carpenter, André Olivier.
From the Department of Neurology and Neurosurgery, McGill University and the Montreal Neurological Hospital and Institute
Running title: Atrophy of amygdala and ictal fear.
Key Words: MRI, volumetric study, mesial temporal sclerosis, amygdala, hippocampus, ictal fear, seizures
Correspondence to : Frederick Andermann, MD FRCP(C) Montreal Neurological Hospital and Institute 3801 University St.; Montreal, Quebec Canada H3A 2B4 - Tel. (514) 398-1976 FAX 514-398-8540
Cendes et al 1 ABSTRACT Viscerosensory and affective manifestations, are often elicited by temporal lobe seizure discharges. They have been reproduced by amygdaloid stimulation in awake patients during stereotaxic exploration or neurosurgical procedures. They are not exclusively reproduced by stimulation of the amygdala, though most commonly they are evoked from it. Ictal fear is frequently, but not invariably, associated with a rising epigastric sensation, palpitations, mydriasis and pallor. We studied 50 patients (mean age 33 yrs) with intractable temporal lobe epilepsy (TLE): MRI volumetric measurements of amygdala and hippocampus were performed using a protocol previously described by our group. All patients had extensive EEG investigation and at least two seizures recorded by video-EEG monitoring. Seventeen patients (34%) had a clear history of fear accompanied by a rising epigastric sensation as the initial manifestation of their habitual attacks. The amygdala volumes in this group were significantly (p < 0.0001) smaller (mean 2131.6mm3) compared to the volumes of the 33 patients without these symptoms (mean 2561.5mm3). Both patient groups had smaller mean amygdala volumes compared to normal controls (mean 2828.2mm3). Postoperative pathology correlated well with volumetric atrophy. In addition, we found that patients with more pronounced amygdaloid atrophy more commonly had prolonged febrile convulsions in early childhood and also more frequently secondarily generalized seizures. Results support the finding that ictal fear is related to pathology of the amygdala and that it, like the hippocampus, is an important substrate of TLE.
Cendes et al 2 It has long been known that the mesial structures of the temporal lobe which include the amygdala and the hippocampus play an important role in the pathogenesis of seizures in temporal lobe epilepsy (TLE) (Gloor, 1992; Penfield, 1954; Feindel, Penfield, 1954; Ben-Ari, 1981; Bancaud et al.1966; Wieser, 1983). The specific involvement of the amygdala in seizures of temporal lobe origin was put in evidence for the first time in 1954 by Feindel and Penfield's observations during stimulation in the amygdaloid region of patients undergoing surgery for TLE (Feindel & Penfield, 1954; Penfield & Jasper, 1954). Viscerosensory and affective manifestations are often activated by temporal lobe seizure discharge. They have been reproduced by amygdaloid stimulation in awake patients during stereotaxic exploration or neurosurgical procedures. They are not exclusively reproduced by stimulation of the amygdala, though most commonly they are evoked from it (Gloor, 1992; Fish et al.1993). Ictal fear is frequently, but not invariably, associated with a rising epigastric sensation, palpitation, mydriasis and pallor. This may be associated with a fearful hallucination, a frightful memory flashback, or both (Gloor, 1972,1992; Gloor et al.1981). The usefulness of quantitative MRI-based volume measurements of the hippocampus in improving detection of unilateral atrophy in persons with epilepsy, has recently been reported (Cascino et al.1991; Jack et al.1990,1992; Cook et al.1992; Cendes et al.1993a,1993b). The atrophy demonstrated by volumetric study showed good correlation with MTS and neuronal loss in the postoperative histopathology (Cascino et al.1991; Jack et al.1990,1992; Cook et al.1992; Cendes et al.1993a,1993b). We performed this investigation, using MRI volumetric studies, in an attempt to
Cendes et al 3 determine if a relationship exists between the degree of atrophy of the amygdala or the hippocampus and the presence of ictal fear.
PATIENTS AND METHODS Patients We studied 50 consecutive patients with TLE (mean age 33 years) investigated at the Montreal Neurological Institute and Hospital. There were 30 women and 20 men. The type and site of seizure onset were determined by a comprehensive evaluation including a detailed history and neurological examination, review of medical records, neuropsychological evaluation, routine MRI studies and extensive EEG investigation consisting of prolonged EEGs with sphenoidal electrodes, using the 10-20 International system for electrode placement, and intensive monitoring with video-EEG telemetry to record seizures. Each patient had at least two of his/her habitual seizures recorded, except for five patients. Nine of the 50 patients underwent intracranial EEG recordings with stereotaxically implanted depth electrodes because the extracranial EEG recordings and other investigations did not provide clear localization or lateralization of seizure onset. The estimation of frequency of secondary generalized seizures was based on review of medical records, seizure calendars and specific questioning of the patient and family members. The evaluation of history of prolonged febrile convulsions (PFC) in early childhood was based on detailed accounts from parents and other relatives as well as on a review of the patient's medical records of early childhood hospitalizations. PFCs were defined as lasting more than 30 minutes.
Cendes et al 4 MRI acquisition for volumetric study The MRI acquisition protocol for volumetric measurements consisted of a scout sequence to ensure proper position of the subject's head and to define the plane of the lateral sulcus on sagittal images. Thirty-two slices in coronal views perpendicular to this plane were then obtained. For the first 40 patients we used a 3D gradient fast field echo (FFE) sequence with 3 mm interleaved sections, a 75/16/2 (TR/TE/ number of signal averages) pulse sequence, a matrix size of 256x256, with a 250 mm field of view and a 60 degree flip angle. For the 10 most recently studied patients we acquired coronal T1 weighted images using 2DF with inversion recovery and 2000/228/21 TR/TI/TE; also using 3mm-tick contiguous coronal slices. This type of acquisition produces higher quality images with better grey/white matter contrast. This MRI acquisition protocol provide us with 6 ± 1 cross-sectional area for amygdala and 15 ± 1 for hippocampal formation measurements.
Volumetric Analysis The images were transferred to a Sun SPARC work station (Sun Microsystems, Mountain View, Cal.). Volumetric measurements were performed with an interactive software program. The regions of interest (ROI) were outlined using a manual contouring editing function. Once the outline had been defined, the slice volume was calculated automatically by the computer program and expressed in cubic millimetres. Anatomic guidelines for outlining the amygdala and hippocampus followed a specific protocol previously described in details by Watson et al (1992).
Cendes et al 5 Volumes of the amygdala and hippocampus of each patient were obtained and compared with values from a control group consisting of 30 healthy volunteers (mean age 32.6 y., range 20-50 y.). Significant volume reduction of each structure, was defined as values 2 SD below the corresponding mean value in the control group. We used a ratio to correct for individual variance of the head size. This consisted of a) dividing the mean 'total brain volume' of the controls by the patient 'brain volume'; b) in each patient, this ratio was then multiplied by the calculated volume of each structure (i.e. amygdala and hippocampus) (Cendes et al.1993b). This correction for 'brain volume' assumes a linear relationship between volumes of amygdala and hippocampal formation, which we were able to demonstrate in our group of 30 normal volunteers (regression analysis, F=17.7, p
