Epilepsy syndromes associated with hypothalamic hamartomas
Article Outline
Summary
Purpose
Hypothalamic hamartoma (HH) related epilepsy presents with gelastic seizures (GS), other seizure types and cognitive deterioration. Although seizure origin in GS has been well established, non-GS are poorly characterized. Their relationship with the HH and cognitive deterioration remains poorly understood. We analyzed seizure type, spread pattern in non-GS and their relationship with the epileptic syndrome in HH.
Methods
We documented all current seizure types in six adult patients with HH-epilepsy with video-EEG monitoring, characterized clinical–electrographic features of gelastic and non-gelastic seizures and correlated these findings with cognitive profile, as well as MRI and ictal SPECT data.
Results
Only four seizure types were seen: GS, complex partial (CPS), tonic seizures (TS) and secondarily generalized tonic–clonic seizures (sGTC). An individual patient presented either CPS or TS, but not both. GS progressed to CPS or TS, but not both. Ictal patterns in GS/TS and in GS/CPS overlapped, suggesting ictal spread from the HH to other cortical regions. Ictal SPECT patterns also showed GS/TS overlap. Patients with GS–CPS presented a more benign profile with preserved cognition and clinical-EEG features of temporal lobe epilepsy. Patients with GS–TS had clinical-EEG features of symptomatic generalized epilepsy, including mental deterioration.
Conclusions
Video-EEG and ictal SPECT findings suggest that all seizures in HH-related epilepsy originate in the HH, with two clinical epilepsy syndromes: one resembling temporal lobe epilepsy and a more catastrophic syndrome, with features of a symptomatic generalized epilepsy. The epilepsy syndrome may be determined by HH size or by seizure spread pattern.
Keywords: Hypothalamic hamartoma, Epilepsy, Cognition, Video-EEG, MRI, SPECT
Introduction
Gelastic seizures (GS) are the hallmark of hypothalamic hamartoma (HH) related epilepsy.1 Other seizure types may also occur, which are often refractory to medical treatment.2 Most video-EEG studies of HH related seizures focus on GS,3 which are not the most disabling seizure type.4 Video-EEG features of non-GS have received less attention. A more precise description of the multiple seizure types, with detailed ictal EEG correlation should be useful not only to characterize these seizures, but also to allow a better understanding of the underlying mechanisms of seizure origin and spread patterns in non-GS and to clarify the role of the cerebral cortex structures in seizure origin.
Most studies of HH related epilepsy included both children and adults.2, 5, 6, 7 However, seizure types in children may vary in an age-dependent manner.2, 8 Hence, studying a more homogeneous series of adult patients with stable seizure types may offer an advantage to determine seizure origin and spread.
We studied video-EEG features of GS and non-GS in a series of adult patients with HH and correlated seizure patterns, as well as ictal SPECT, to discuss possible underlying seizure origin and spread patterns and epilepsy syndromes in HH-epilepsy.
Patients and methods
Six adult patients with refractory epilepsy and MRI documented HH underwent continuous video-EEG monitoring with a 64-channel BMSI-Nicolet 5000 video-EEG equipment (Madison, WI), with standard electrode placement (10–20 system) and additional anterior temporal and subtemporal electrodes (total of 27 electrodes) to record all current seizure types. We analyzed clinical features: seizure types and number of seizures, seizure patterns, i.e. how seizures evolved to different seizure types for each seizure, and ictal EEG correlates of all seizures.
All patients had undergone 1.5
T MRI scans of the brain, consisting of T1-weighted sequences in coronal, sagittal and axial views and additional T2-weighted and FLAIR sequences, including at least one coronal view. Brain MRIs were retrospectively reviewed in all cases by a neuroradiologist who characterized HH features according to size, lateralization and localization (middle or posterior).6
When possible, interictal and ictal single photon emission computed tomography (SPECT) were performed with an intravenous injection of approximately 740MBq (20mCi) of 99mTc-ECD followed by image acquisition with a dual head gamma camera with dedicated collimator for brain studies (fan beam), E.CAM (Siemens, Hoffman Estate Il). Rotation time was determined through counts per frame (at least 100,000 counts per frame) and 128 projections were acquired. SPECT images were processed on a workstation E-soft (Siemens) with Butterworth filter (with 0.75 of cutoff and order 5). Reconstruction yielded 4.8mm voxels with a 128×128 matrix and 128 slices. In-plane spatial resolution was 10.6/6.7mm full width at half maximum (FWHM) in the center of view. SPECT studies were reviewed by a nuclear medicine specialist who characterized increased regional blood flow abnormalities.
Lastly, we analyzed combined data of seizure types and spread patterns, MRI as well as ictal and interictal SPECT findings to define epilepsy syndromes in these patients.
Results
We studied six adult patients with HH (Table 1). Four were men. Ages ranged from 18 to 39 years (mean, 26.8; median, 25). Only patient 6 had a history of precocious puberty, at age 2. Age at epilepsy onset ranged from newborn period to 8 years. All patients had medically refractory seizures and stable seizure types for long periods (from 3 up to 24 years without new seizure types) before the video-EEG studies.
Table 1. Gender, age, epilepsy onset and cognitive function of six patients with hypothalamic hamatoma and epilepsy
| Patient | Gender | Age (years) | Epilepsy onset (age) | Cognitive function (IQ)* |
|---|---|---|---|---|
| 1 | F | 38 | 5 years | Normal (126) |
| 2 | F | 23 | 2 years | Normal (IQ not measured) |
| 3 | M | 18 | 6 months | Severe impairment (22) |
| 4 | M | 39 | 8 years | Severe impairment (11) |
| 5 | M | 18 | 5 years | Mild impairment (64) |
| 6 | M | 25 | Newborn | Severe impairment (15) |
All patients had sessile HHs, measuring (largest diameter) 7–33mm, with volumes ranging from 0.1 to 9.9cm3 (Fig. 1). MRI showed right lateralization in four patients and left in two. Four HHs were posterior, one was middle and one was unclassifiable6 (Table 2).
Figure 1.
T1 weighted MRI of HH in all cases. (A) Patient 1, coronal view. (B) Patient 2, sagittal view. (C) Patient 3, sagittal view. (D) Patient 4, coronal view. (E) Patient 5, coronal view. (F) Patient 6, sagittal view. Arrows point to HH, in each case.
Table 2. MRI findings in six patients with hypothalamic hamartoma and epilepsy
| Patient | Side | Location6 | Dimensions (cm) | Volume (cm3) |
|---|---|---|---|---|
| 1 | Left | Middle | 0.7×0.7×0.7 | 0.1 |
| 2 | Right | Middle | 0.7×1.2×1.3 | 0.5 |
| 3 | Right | Not classifiable | 1.3×1.4×1.5 | 1.4 |
| 4 | Right | Posterior | 1.5×1.5×1.0 | 1.1 |
| 5 | Right | Middle | 1.1×1.3×1.2 | 0.8 |
| 6 | Left | Middle | 2.0×2.9×3.3 | 9.9 |
Patients 1 and 2 presented normal cognitive function and had the smallest HHs. The other four patients had mild to severe cognitive function impairment (Table 1, Table 2).
All patients underwent continuous video-EEG monitoring, lasting 3–7 days. All current seizure types were recorded in all cases. Four seizure types were seen: gelastic seizures (GS), complex partial (CPS), tonic (TS) and secondarily generalized tonic–clonic seizures (sGTC).
GS consisted of episodes of unprovoked smile and laughter-like behavior. Consciousness could be impaired during GS. GS were present in five patients. In a given patient, GS could evolve either to CPS or to TS, but not to both. Patients 1 and 2 presented GS occurring isolatedly or evolving to CPS, whereas patients 5 and 6 had isolated GS or GS that evolved into TS. Dacrystic seizures or autonomic phenomena were not seen in this patient series.
Ictal EEG of GS showed three distinct patterns. The most common pattern in GS was of no ictal changes. This pattern was seen in all patients with GS, except in patient 2. A diffuse ictal pattern (diffuse paroxysmal fast activity or diffuse attenuation) was seen during GS in two patients (patients 5 and 6), who presented GS that occurred in isolation or that later evolved to TS. An identical pattern was seen in TS in these patients. A focal ictal onset in the temporal region, with rhythmic or semi-rhythmic theta activity was seen during GS in two patients (patients 1 and 2), who presented isolated GS or GS evolving to CPS. This pattern was the same ictal pattern seen during CPS in these patients.
TS were characterized by asymmetric bilateral arm elevation and abduction, forceful anterior head flexion or lateral head deviation, accompanied or not by upward conjugate eye deviation. Consciousness was always impaired in this seizure type. TS occurred randomly in the sleep–wake cycle in all patients, without seizure predominance in sleep or after awakening. This seizure type has been described in HH.9 Four patients had TS (patients 3, 4, 5 and 6). In three patients (patients 3, 5 and 6) GS could precede or occur concomitantly with the TS. Patient 4 had isolated TS. Ictal EEG during TS showed diffuse attenuation or paroxysmal fast activity (Fig. 2), followed by right frontal ictal activity in one patient. In this series, we did not see the “evolving generalized fast rhythmic activity”, described in TS, CPS or atypical absences occurring in association with HH.2
Figure 2.
Ictal EEG during a tonic seizure (patient 6) showing a high-amplitude, diffuse slow (delta) wave, followed by diffuse attenuation of the background activity, and superimposed fast activity (beta) more prominently noted in the fronto-temporal electrodes. Referential montage, with average reference. The arrow points to the clinical and EEG onset.
The remaining two patients (patients 1 and 2) had CPS, consisting of episodes of impaired consciousness with manual or orofacial automatisms. Ictal EEG showed unilateral rhythmic theta activity in the temporal and subtemporal electrodes in one patient (Fig. 3) and in the fronto-temporal electrodes in another. In both patients, GS could precede or occur concomitantly with the CPS.
Figure 3.
Ictal EEG during a complex partial seizure (patient 1) showing sharply contoured (theta) activity in the left temporal and subtemporal electrodes. Referential montage, with average reference. The arrow points to the clinical and EEG onset.
The number of seizures recorded, as well as seizure types, ictal EEG patterns for each seizure type and evolving seizure type patterns for each patient are shown in Table 3.
Table 3. Video-EEG findings in six adult patients with hypothalamic hamartoma and epilepsy
| Patient | Seizure type (number of recorded seizures) | Ictal EEG | Seizure patterns |
|---|---|---|---|
| 1 | GS (6) | No changes or left temporal, theta activity | iGS, GS→CPS, CPS→sGTC |
| CPS (5) | Left temporal, theta activity | ||
| sGTC (4) | Generalized | ||
| 2 | GS (2) | Bilateral frontal-temporal activity, predominantly right | iGS, GS→CPS, iCPS |
| CPS (2) | Bilateral frontal-temporal activity, predominantly right | ||
| 3 | GS (2 clusters) | No changes or diffuse desynchronization with paroxysmal fast activity | iGS, iTS, TS→sGTC |
| TS (6) | No changes or diffuse desynchronization and paroxysmal fast, followed by 4Hz spike-and-wave in the anterior regions | ||
| sGTC (2) | Generalized | ||
| 4 | TS (7 and 1 cluster) | Diffuse paroxysmal fast activity followed by desynchronization | iTS |
| 5 | GS (17) | No changes or diffuse attenuation and paroxysmal fast activity, predominantly right frontal-central-temporal | iGS, GS→TS, GS→TS→sGTC |
| TS (11) | Diffuse attenuation and paroxysmal fast activity, predominantly right frontal-central-temporal; one seizure right temporal theta-alpha activity | ||
| sGTC (3) | Generalized | ||
| 6 | GS (9) | No changes or diffuse desynchronization | iGS, GS→TS |
| TS (8) | No changes or diffuse desynchronization | ||
| sGTC (1) | Generalized | GS→TS→sGTC | |
The two patients with CPS had the smallest HH and preserved cognitive functioning: both attended college and worked independently, despite having medically refractory seizures. HH side correlated with EEG lateralization during CPS seizures. The four patients with TS had significant intellectual impairment. Three were severely retarded (IQ ranging from 11 to 22) and one had mild mental retardation (IQ of 64). Ictal EEG was diffuse, showing no lateralization in three of these patients (patients 3, 4 and 6). In patient 5, a diffuse pattern was seen, predominating on the right, in concordance with HH lateralization on MRI.
Ictal SPECT was obtained in five patients. Increased regional blood flow was seen in two. In the other three patients no ictal changes were seen (Table 4). One study, obtained during a CPS (patient 2) showed right temporal hyperperfusion. Concomitant ictal EEG showed focal ictal activity in the right frontal-temporal region. The other abnormal study, performed during a GS evolving to a TS (patient 6), showed increased cerebral blood flow in the hamartoma, left frontal region and in the right cerebellum (Fig. 4). Concomitant ictal EEG showed only movement and muscle artifacts.
Table 4. Ictal SPECT findings in six patients with hypothalamic hamartoma and epilepsy
| Patient | Seizure type–ictal SPECT injection | Ictal EEG during SPECT injection | Ictal SPECT (hyperperfusion) |
|---|---|---|---|
| 1 | CPS→sGTC | Left temporal | No changes |
| 2 | CPS | Bilateral frontal-central-temporal, predominantly right | Right temporal |
| 3 | Not performed | Not performed | Not performed |
| 4 | TS | Diffuse paroxysmal fast activity followed by desynchronization | No changes |
| 5 | GS→TS | Diffuse attenuation and paroxysmal fast activity, predominantly right frontal-central-temporal | No changes |
| 6 | GS→TS | Artifacts | Left frontal and hamartoma |
Figure 4.
Interictal and ictal SPECT obtained during a gelastic seizure evolving to a tonic seizure (patient 6). Note ictal increased flow in the HH (white arrow), left frontal lobe (yellow arrow). Increased flow was also seen in the right cerebellum (not shown). MRI to provide anatomical correlation: sagital T1 weighted image, axial FLAIR (arrows point to HH).
Discussion
Seizure types
Multiple seizure types are reported in HH-epilepsy, but most studies have not analyzed in detail the clinical and EEG features of non-gelastic seizures. In this series, contrary to what is commonly referred in the literature, we found a surprisingly limited number of seizures types for each individual patient. Other than GS, only two major seizure types were seen: CPS and TS. Seizure spread patterns were limited to two types: GS could progress either to a CPS or to TS. An individual patient presented with either CPS or TS, but not both. Both seizure types (CPS and TS) could evolve to secondary generalized seizures.
Tonic seizures are classically described as symmetrical muscular contraction, chiefly in the axial muscles and proximal limbs with loss of consciousness and an ictal EEG correlate of generalized rhythmic fast activity. Asymmetrical postural seizures (also called focal tonic seizures or supplementary motor seizures) are described as tonic posturing of the limbs in an asymmetrical fashion (e.g. ‘fencing posture’), during which consciousness may not be impaired. EEG characteristically shows focal epileptic activity in the frontocentral region. Clinical and even electroencephalografic distinction between tonic seizures and asymmetrical postural seizures is often difficult. Tonic seizures frequently present variable degrees of asymmetrical limb contraction, as well as head and eye deviation. Likewise, while ictal EEG in tonic seizures may present as focal epileptic activity in the frontal region, ictal EEG in asymmetrical postural seizures may show generalized activity.10 We considered that patients 3, 4, 5 and 6 had tonic seizures with a certain degree of asymmetry in clinical, electroencephalographic and SPECT findings. Our patients did not show features of focal tonic seizures, such as unilateral tonic activity or asymmetrical posturing, resembling the fencing posture or preserved consciousness during tonic seizures.
In a detailed literature review, similar seizure types have been described in different ways. Many studies describe this seizure pattern simply as drop attacks or atonic seizures.1, 2, 4, 5, 8, 11 Other studies suggest a tonic seizure type.1, 3, 4, 5, 6 In most studies with a more detailed seizure description, focal features become more apparent: head turning,3 tonic seizures with head and eye rotation, arm elevation, as well as left eyelid and mouth clonic jerking, producing a fall;6 head rotation and drop; drop attacks with head turning, four limb postural seizures suggesting supplementary motor area involvement,6 behavioral arrest and consciousness impairment with left tonic head and body deviation thought to arise from the right frontal lobe.8 In a stereo-EEG study, seizures were described as sudden backward fall with eye deviation to the right, sometimes associated with right arm tonic posturing, right or left head turning and contractions of the corners of the mouth. Electrically, a low voltage fast activity was recorded, particularly over the right cingulus, second frontal gyrus, right and left supplementary motor area, right precentral opercular region sparing the lesion but involving posterior hypothalamic structures only over their posterior parts.9 It is conceivable that all these seemingly different seizure types represent subtle variations on a single seizure type, with tonic seizure features.
Ictal EEG/ictal SPECT patterns
Frequently, an ictal EEG correlate was lacking during GS. This is in agreement with ictal onset in a deep-seated structure. Ictal activity during GS may not be captured by scalp electrodes, but can be seen in the HH when depth electrode recording is used.9
Despite the fact that EEG during GS more commonly did not show ictal abnormalities, we noted ictal EEG abnormalities during some GS. In two patients (patients 1 and 2), rhythmic theta activity was seen during a GS. These patients also had CPS with this same ictal correlate. In three other patients, a diffuse attenuation pattern or diffuse paroxysmal activity was seen during GS (patients 3, 5 and 6). In patient 5, ictal activity during a GS was diffuse, predominating in the right frontal-temporal region. These patients had GS evolving to TS, with an identical ictal EEG pattern. The finding of overlapping clinical and ictal EEG patterns in isolated GS and GS evolving to other seizure types lend further support to the idea that other seizure types seen in these patients represent seizure activity spreading from the HH. In patients 1 and 2, CPS could evolve from a GS and a similar ictal EEG pattern was seen in both isolated GS and in GS evolving to CPS, suggesting that CPS in these patients may represent ictal activity spread from the HH to the temporal lobe. In patients 3, 5 and 6, who presented with GS progressing to TS, ictal EEG showed a diffuse ictal pattern (paroxysmal fast activity or diffuse attenuation), which could be asymmetrical (patient 5). This pattern was seen in both isolated GS and in GS evolving to TS.
The ictal pattern of diffuse desynchronization or paroxysmal fast activity can be seen in either primary generalized tonic seizures or in focal seizures originating in midline structures, such as the supplementary motor area. Tonic seizures are thought to represent primary generalized seizures. Underlying mechanisms of seizure onset and spread in TS remain incompletely understood. In our cases, the overlap of clinical and ictal EEG data during isolated GS and in GS evolving to TS suggests that TS represent ictal spread from the HH to more cranial midline structures.
We find further support to this notion when we analyze ictal SPECT data. In patient 2, an ictal SPECT injected during a CPS, with an ictal EEG correlate of bilateral fronto-temporal rhythmic theta activity, predominantly right, showed right temporal lobe increased flow. More compelling data is the finding, in patient 6, of increased flow in both the hamartoma and in the left frontal region when ictal SPECT was injected in a GS evolving to a TS. Ictal EEG during this seizure showed no ictal changes, obscured by artifacts. In patient 6, the clinical pattern of progression (GS→TS), as well as SPECT findings strongly suggest that in this setting a TS represents ictal activity spread from the hamartoma to more cranial structures, probably involving deep midline structures and the frontal lobe.
The hypothalamus is anatomically connected to the anterior thalamus through the mammillothalamic tracts and the spread of epileptic activity could impair the corticothalamic loops.12 It has been also suggested that “an interaction between cortical and subcortical motor areas may be involved in the production of all tonic seizures, generalized, bilateral asymmetrical and focal”.10
Epileptic spasms have been described in association with HH.2, 7 We did not observe epileptic spasms in our series. This may be due to the fact that our series only included adult patients with stable seizure types. Infantile and epileptic spasms are age-dependent seizure types and are closely related to tonic seizures and symptomatic generalized epilepsy.10
MRI correlation
In a previous study of seven patients with HH-epilepsy, the authors classified HHs as middle or posterior, according to preservation of increased signal halo on T1-weighted images of the mammilary bodies.6 The authors suggested that posterior HHs were associated with temporal lobe seizures, while middle HH were associated with frontal lobe seizures. Though we have not performed specific MRI studies as described by that study, we have not found association between HH/mamillary body relationship and seizure type or spread pattern. Two patients with temporal lobe seizures had middle HH and of the four patients with frontal lobe seizures, two had middle HH, one patient had posterior HH and in the remaining patient, the HH was unclassifiable.
In our series, patients with smaller HH had CPS, while patients with larger lesions had features of a symptomatic generalized epilepsy. The association between smaller HH and more preserved cognition has been reported.7, 13
Ictal EEG correlated with HH-MRI lateralization in the two patients with clinical and EEG features of temporal lobe epilepsy and in one patient with TS and diffuse, but assymetrical EEG findings.
Clinical syndromes
The findings described above strongly argue that all seizure types (CPS and TS) seen in these patients with hypothalamic hamartomas originate from ictal activity spread from the hamartoma to other subcortical and cortical regions, also arguing against the occurrence of independent cortical foci in HH. These ideas are further supported by data showing that isolated HH resection may resolve all seizure types in epilepsy associated with hypothalamic hamartomas,4, 14 while limited cortical resections in HH have uniformly failed to attain seizure control in patients with HH.11, 15 A “running-down phenomenon” or delayed improvement in seizure control or even complete seizure cessation has been recognized after HH resection in some patients. Rather than arguing against seizure origin in the HH, this phenomenon suggests the occurrence of reversible cortical changes, leading to seizures which may resolve with HH resection. It has been suggested that seizure spread may lead to incomplete secondary epileptogenesis, which is reversed after a variable period of time (months to years) following HH removal.4, 15 The mechanisms underlying this phenomenon remain currently incompletely understood.
In our study, the patients that presented with clinical and EEG features suggestive of temporal lobe epilepsy had a more benign clinical profile, with more preserved cognition and better social and employment status, despite medically refractory seizures. It has been suggested that ‘pseudotemporal lobe seizures’ are caused by epileptic activity in the HH that reach the temporal lobe through hypothalamic-amygdala connections.12 Patients with this clinical syndrome corresponded to the smaller hamartomas. It is unclear whether the smaller hamartoma size or the seizure spread pattern with clinical features of temporal lobe epilepsy, or both, is the determinant factor of a more benign clinical profile.
Patients with TS had significant cognitive impairment and more severe epilepsy. Clinical presentation of these patients is akin to the symptomatic generalized epilepsies12 or to an epileptic encephalopathy syndrome, with multiple seizure types, including TS and cognitive decline. The underlying mechanism leading to cognitive deterioration in these epilepsy syndromes remains poorly understood. Patients with larger HH and TS presented more severe epilepsy and cognitive deterioration. It is possible that the spread pattern, involving midline structures and hamartoma size play important roles in the more severe clinical presentation.
We suggest that seizures originate in the HH, with two major epilepsy syndromes being discerned in HH related epilepsy. One syndrome consists of patients of more favorable profile, with GS associated with clinical and EEG features of medically refractory temporal lobe epilepsy, with preserved cognition, which was associated with smaller lesions. The other group consists of patients with a catastrophic epilepsy syndrome with GS, associated with other seizure types that indicate a symptomatic (or probably symptomatic) generalized epilepsy syndrome associated with cognitive decline. The underlying factors that determine whether patient will evolve into one of the patterns are poorly understood. It is possible that a vertical spread pattern, involving deep midline structures and frontal lobe structures, including the supplementary motor area is associated with more severe epilepsy syndrome and cognitive deterioration, suggestive of a symptomatic generalized epilepsy and that lateral spread pattern to the temporal lobe determines a more benign syndrome, of a focal “pseudo-temporal” epilepsy. It is also unclear if a presentation with clinical features of an extratemporal focal epilepsy without cognitive deterioration, reflecting ictal spread patterns to a different cortical region, not involving deep midline structures, may occur.
Acknowledgements
Ethics Committee approval: This study has been approved by the Ethics Committee of Hospital das Clinicas da Faculdade de Medicina, Universidade of São Paulo, São Paulo, Brazil (CAPPesq). Research protocol number 246/03. Funding: The study was supported by FAPESP (The State of São Paulo, Brazil Research Foundation) Scientific Initiation Grant Number: 03/12920-5.
References
- . Hypothalamic hamartomas and ictal laughter: evolution of a characteristic epileptic syndrome and diagnostic value of magnetic resonance imaging. Ann Neurol. 1988;23:429–439
- The clinical spectrum of epilepsy in children and adults with hypothalamic hamartoma. Epilepsia. 2003;44:1310–1319
- Intrinsic epileptogenesis of hypothalamic hamartomas in gelastic epilepsy. Ann Neurol. 1997;42:60–67
- . Generalized epilepsy in hypothalamic hamartoma: evolution and postoperative resolution. Neurology. 2003;60:762–767
- . Hypothalamic hamartoma with gelastic seizures in Swedish children and adolescents. Eur J Paediatr Neurol. 2004;8:35–44
- . Different electroclinical manifestations of the epilepsy associated with hamartomas connecting to the middle or posterior hypothalamus. Epilepsia. 2003;44:1191–1195
- . The clinical spectrum and natural history of gelastic epilepsy-hypothalamic hamartoma syndrome. Seizure. 2005;14:232–239
- Hypothalamic hamartomas: seven cases and review of the literature. Epilepsy Behav. 2003;4:246–258
- Role of the hypothalamic hamartoma in the genesis of gelastic fits (a video-stereo-EEG study). Electroencephalogr Clin Neurophysiol. 1995;95:154–160
- . Tonic seizures. In: Lüders HO, Noachtar S editor. Epileptic seizures: pathophysiology and clinical semiology. Philadelphia, PA: Churchill Livingstone; 2000;p. 389–411
- Gelastic seizures and hypothalamic hamartomas: evaluation of patients undergoing chronic intracranial EEG monitoring and outcome of surgical treatment. Neurology. 1993;43:747–750
- . Human epileptogenesis and hypothalamic hamartomas: new lessons from an experiment of nature. Epilepsia. 1997;38:1–3
- . “Pressure to laugh”: an unusual epileptic symptom associated with small hypothalamic hamartomas. Neurology. 2000;54:971–973
- . Resective surgery for hypothalamic hamartoma. Epileptic Disord. 2003;5:281–286
- . The hypothalamic hamartoma: a model of subcortical epileptogenesis and encephalopathy. Semin Pediatr Neurol. 2005;12:119–131
PII: S1059-1311(06)00198-1
doi:10.1016/j.seizure.2006.10.008
© 2006 British Epilepsy Association. Published by Elsevier Inc. All rights reserved.
