Gelastic seizures and the anteromesial frontal lobe: A case report and review of intracranial EEG recording and electrocortical stimulation case studies


Address correspondence to Kanjana Unnwongse, Cleveland Clinic Epilepsy Center, Neurological Institute, S-51, 9500 Euclid Avenue, Cleveland, OH 44195, U.S.A. E-mail:


Symptomatogenic areas for ictal laughter have been described in the frontal and temporal lobes. Within the frontal lobe, gelastic seizures have been recorded from the cingulate gyrus. Electrocortical stimulation of the cingulate gyrus as well as the superior frontal gyrus induced laughter. We describe a patient whose gelastic seizures were associated with electrographic ictal activity in the mesial aspect of the right anterior frontal gyrus. The symptomatogenic area for ictal laughter in the frontal lobe may reside in the superior frontal gyrus.

Gelastic seizures are characterized by unnatural forceful laughter with or without mirth. They have been associated with hypothalamic hamartomas, as well as cortical foci in the frontal (Kovac et al., 2009), temporal (Loiseau et al., 1971), and parietooccipital lobe (Kawakami et al., 2000). Very few studies have employed intracranial electroencephalography (EEG) and electrocortical stimulation to further localize the symptomatogenic zone of laughter in the frontal lobe. Herein, we present a patient whose pharmacoresistant gelastic seizures were demonstrated to arise from the right anterior mesial frontal gyrus during intracranial video-EEG recording.

Case Report

A 66-year-old right-handed man was referred for pharmacoresistant epilepsy. The patient had stereotypic seizures since childhood. Seizures were characterized by a sudden hysterical laugh accompanied by a frozen facial expression and unresponsiveness, lasting less than one minute. They occurred almost exclusively in the early morning hours, both from the awake and sleep state. The patient had no risk factors for epilepsy other than a prolonged delivery due to breech presentation. He graduated from high school and worked as an electrician for more than 30 years. During the 2-year period preceding the referral, his seizure frequency and morbidity had increased. He had been hospitalized several times for status epilepticus. He had also sustained an open fracture of his right thumb during a seizure.

Scalp video-EEG monitoring revealed normal wake and sleep patterns with no interictal abnormalities. After antiepileptic drug (AED) reduction, more than 20 stereotypical seizures were recorded, characterized by sudden laughing or giggling, sometimes preceded by an exclamation “Jesus Christ.” This was followed by flexion of the hips, raising the legs off the bed, and nonforceful turning to the right. Ictal EEG consistently showed diffuse electromyography (EMG) and movement artifacts with no localizable seizure pattern. A 3T brain magnetic resonance imaging (MRI) demonstrated an area of increased T2 signal hyperintensity extending from the right superior frontal gyrus to the cingulate gyrus (Fig. 1A), corresponding to an area of hypometabolism on positron emission tomography (PET) imaging (Fig. 1B). These findings were interpreted as a transmantle cortical dysplasia. An ictal single photon emission computed tomography (SPECT) study injected 12 s after seizure onset demonstrated an area of hyperperfusion in the right anterior frontal region (Fig. 1C). Functional MRI and intracarotid methohexital (WADA) test showed left hemispheric language dominance. An intracranial evaluation was performed using subdural electrodes covering the right mesial, lateral, and orbitofrontal regions. Interictal runs of spikes were observed every 10–20 s, over 95% from the area overlying the lesion and the remaining from the orbitofrontal area. Twelve habitual seizures were recorded. EEG onset was characterized by high frequency, low voltage activity arising from the anterior mesial aspect of the right superior frontal gyrus, approximately 1 cm above the cingulate gyrus (Fig. 1). Electrocortical stimulation was performed to delineate the primary motor cortex. Responses characteristic for the supplementary motor area such as negative motor phenomena and tonic contractions of the left hemibody or left extremities (Fig. 2) were acquired 2.5 cm posterior to the seizure onset. Stimulation was aborted before assessing the seizure onset zone due to recurrent unprovoked seizures. The patient underwent a right frontal resection that included the superior and middle frontal gyri as well as the cingulate and orbitofrontal gyri. Histopathology demonstrated cortical dysplasia (Taylor type IIB). He has been entirely seizure free for 1 year.

Figure 1.

Left upper insert: Position of subdural grid contacts projected onto surface-rendered magnetic resonance imaging (MRI) of the right lateral and mesial frontoparietal aspect. Interictal spikes and ictal onset with spreading patterns of 12 seizures are summarized. Findings of intracranial electroencephalography (EEG) recording, Left: Interictal spikes were limited to contacts SE 5, 11, and 12 in the vicinity of the MRI abnormality (green circles on grid). Right upper: Habitual seizure video snapshots correlated with intracranial EEG, time points of individual captions indicated by black arrow. Right: Ictal EEG onset (A) characterized by low amplitude paroxysmal fast activity restricted to contacts SE 5, 6, and 11 (red circles on grid), overlying the right anteromesial frontal region. Two seconds later this activity was observed in contacts SB 10, SC 11, and 23 (orange circles on grid), overlying the lateral and mesial prefrontal regions. Clinical seizure onset 8 s later with smiling (B), followed by laughing (C) and involuntary movements of all extremities (D).


We present a patient with longstanding gelastic seizures due to a transmantle cortical dysplasia extending from the right lateral ventricle into the right superior frontal gyrus. Electrocorticography in this case demonstrated ictal involvement of the right anteromesial superior frontal gyrus in the generation of the ictal laughter, suggesting that this area serves as the symptomatogenic zone. Although other candidate areas, specifically the cingulate gyrus and the temporal lobe, were not evaluated with electrocorticography, the restriction of the ictal activity to three electrodes in the anteromesial superior frontal gyrus, the subsequent hypermotor activity, and, most importantly, postoperative seizure freedom all support frontal lobe origin. Absence of concomitant seizure characteristics such as autonomic symptoms and automatisms argue against anterior cingulate involvement (Devinsky et al., 1995). Although gelastic seizures are most often intracranially recorded from hypothalamic hamartomas (Kahane et al., 2003), ictal laughter can also be generated in the frontal lobe. Within the frontal lobe, the mesial and lateral aspects of the superior frontal gyrus, the cingulate gyrus, as well as the orbitofrontal gyrus, have been associated with gelastic seizures based on intracranial EEG recording and electrocortical stimulation in seven cases (Table 1). The most limited surgical approach resulting in seizure freedom was reported by Chassagnon et al. (2003). Following an invasive evaluation with stereo-EEG, their nonlesional patient became seizure free after radiofrequency stimulation, resulting in two distinct lesions in the left superior frontal and cingulate gyri. Ictal laughter occurred at a relatively late stage during the habitual seizures, suggesting propagation of the ictal activation. Although the typical seizure semiology was reproduced by low-intensity electrocortical stimulation of the contacts that were later used for the radiofrequency ablation, the only spontaneously recorded seizure during the stereo EEG evaluation had no gelastic features. A patient with a cavernous angioma in the left mesial superior frontal gyrus was evaluated using subdural grid and strip electrodes covering the lateral and mesial aspect of the left frontal lobe (Arroyo et al., 1993). At the time of the ictal laughter, four contacts overlying the mesial aspect of the frontal lobe were electrographically involved; however, at that time the seizure had already spread to involve more distant contacts in the anterior lateral frontal lobe and at the frontoparietotemporal junction. Electrocortical stimulation failed to elicit laughter in this patient. A low-grade glioma in the right anterior cingulum caused gelastic seizures in a patient described by Devinsky et al. (1995). Ictal onset was observed from contacts covering the right anterior cingulum, with spread to the right orbitofrontal cortex and the left cingulum within 500 msec. No details were provided with regard to correlation of ictal symptomatology and electrocorticography, or electrocortical stimulation. A patient with a right frontal cortical dysplasia resembling our case was recently reported (Cheung et al., 2007). Although resection of the cortical dysplasia resulted in seizure freedom following an invasive evaluation, no details are provided with regard to the correlation of ictal symptomatology and electrocorticography, or electrocortical stimulation. Lastly, a patient with a pleomorphic xanthoastrocytoma in the left cingulate gyrus, just anterior to the corpus callosum, was rendered free of gelastic seizures following resection of the tumor. Intracranial recordings were not utilized (Mohamed et al., 2007). In the preimaging era, one patient was cured from gelastic seizures after resection of a cystic mass in a similar location (Loiseau et al., 1971; “it looked as if the growth originated from the junction of the corpus callosum with the cingulate cortex, and had invaded both dorsally to the ventricle and rostrally to the mesial cortex.”). Laughter was induced by electrocortical stimulation of five subdural strip contacts covering the left superior frontal gyrus (Fried et al., 1998) and by depth electrode contacts in the right anterior cingulate gyrus (Sperli et al., 2006). It is noteworthy that in the latter case, no depth contacts were located in the superior frontal gyrus. Interestingly, in both these patients, laughter was not part of the patient’s habitual seizure semiology. Therefore, although freedom from gelastic seizures following lesionectomy in the cingulate gyrus suggests that the epileptogenic zone in these cases did indeed reside in this area, electrocorticography from our case in addition to the stimulation data by Fried et al., 1998 suggest that the symptomatogenic zone for the motor manifestations of laughter may reside inside the superior frontal gyrus. Given that epileptogenic lesions have been resected in the cingulate area of either hemisphere without resulting impairment of spontaneous laughter, the “generator” for the motor aspect of laughter is likely represented bilaterally. To our knowledge, no case to date has been reported in whom ictal laughter was reproduced during cortical stimulation. Many variables, such as stimulus duration, frequency, intensity, and anticonvulsant levels affect the results of electrocortical stimulation studies. In our case, stimulation had to be abandoned after definition of the posterior resection margin due to frequent spontaneous seizures. Lastly, one patient with gelastic seizures due to an orbitofrontal oligoastrocytoma has been described (Umeoka et al., 2008). Seizures lasted a few minutes and consisted of sudden motion arrest, unresponsiveness with or without staring, and followed by suspected mirthful laughter and postictal confusion. Intracranial EEG recording showed a leading epileptic discharge at orbitofrontal contacts before spreading to the temporal base where the laughter was observed. Laughter then recurred when the ictal discharge was seen exclusively at the temporal base electrodes. Therefore, the laughter in this case likely represents ictal propagation to the temporal lobe. This view is supported by the fact that stimulation of the parahippocampal gyrus and fusiform gyrus has been demonstrated to elicit laughter (Arroyo et al., 1993). To our knowledge, there is no report of laughter induced by stimulation of the orbitofrontal area.

Table 1.   Evidence for frontal lobe areas involved in generation of laughter, based either on postsurgical seizure freedom in patients with gelastic seizures (with* or without intracranial EEG recordings) or electrocortical stimulation studies eliciting laughter from the frontal area
StudyFrontal lobe areaPathologyLaughter generated byOutcome/follow-up
Spontaneous seizureElectrocortical stimulation
  1. CG, Cingulate gyrus; SFG, superior frontal gyrus; OF, orbitofrontal gyrus; +, positive result; −, negative result.

  2. aBy history only. Laughter did not occur in the seizure recorded during intracranial video EEG monitoring.

Loiseau et al., 1971CG“Astrocytic wall cyst”+Not doneUnknown
Arroyo et al., 1993*CGCavernous angioma+Seizure free/16 month
Devinsky et al., 1995*CGLow grade glioma+Not doneSeizure free/12 month
Fried et al., 1998*SFGNot provided+Unknown
Chassagnon et al., 2003*CGNot provided+aSeizure free/27 month
Sperli et al., 2006*CGNot provided+Unknown
Cheung et al., 2007*SFGCortical dysplasia+Not doneRecurrent aura/12 month
Mohamed et al., 2007CGPleomorphic xanthoastrocytoma+Not doneSeizure free/12 mo.
Umeoka et al., 2008*OFMixed oligoastrocytoma+Not doneSeizure free/33 month


We confirm that we have read the Journal’s position on issues involved in ethical publication and affirm that this report is consistent with those guidelines.


The contributing authors to this article have no conflicts of interest.