- Email: [email protected]
Stereotactic radiofrequency thermocoagulation for hypothalamic hamartoma with intractable gelastic seizures
Epilepsy Research (2007) 76, 15—21
journal homepage: www.elsevier.com/locate/epilepsyres
CLINICAL RESEARCH
Stereotactic radiofrequency thermocoagulation for hypothalamic hamartoma with intractable gelastic seizures Junpei Homma a, Shigeki Kameyama a, Hiroshi Masuda a, Takehiko Ueno a, Ayataka Fujimoto a, Makoto Oishi a,b, Masafumi Fukuda a,b,∗ a
Department of Neurosurgery, Epilepsy Center, Nishi-Niigata Chuo National Hospital, 1-14-1 Masago, Niigata 950-2085, Japan Department of Neurosurgery, Brain Research Institute, University of Niigata, 1-757 Asahimachi-dori, Niigata-City 951-8585, Japan
b
Received 24 December 2006; received in revised form 19 May 2007; accepted 3 June 2007 Available online 23 July 2007
KEYWORDS Gelastic epilepsy; Hypothalamic hamartoma; Radiofrequency thermocoagulation; Stereotactic surgery; Surgical treatment
Summary Management of hypothalamic hamartoma with intractable gelastic epilepsy remains controversial. We have used stereotactic thermocoagulation for treatment of hypothalamic hamartoma with intractable gelastic epilepsy since 1997. Herein, we review our experience in five cases to clarify the usefulness of this treatment. A total of five patients with hypothalamic hamartoma were treated by stereotactic thermocoagulation at our hospital during the period October 1997 through February 2004. In all patients, the hamartoma was less than 10 mm in diameter and was located on the floor of the third ventricle with sessile attachment to the wall. To identify ictal onset, chronic intracranial electroencephalography was performed in three patients with the use of a depth electrode implanted in the hamartoma. Attempts were made to induce gelastic seizure by electrical stimulation of the hamartoma in three patients. After magnetic resonance imaging-guided targeting, radiofrequency thermocoagulation of the boundary between the hamartoma and normal hypothalamus was performed to achieve disconnection effects. Marked reductions in seizure frequency were obtained in all cases, with three patients becoming seizure-free after the procedure. No intraoperative complications occurred except in one patient who experienced acute and transient panidrosis with hot flushes during coagulation. Our results suggest that stereotactic thermocoagulation of hypothalamic hamartoma is an acceptable treatment option for patients with intractable gelastic seizures. © 2007 Elsevier B.V. All rights reserved.
∗
Corresponding author at: Department of Neurosurgery, Brain Research Institute, 1-757 Asahimachi-dori, Niigata-City 951-8585, Japan. Tel.: +81 25 227 0652; fax: +81 25 227 0819. E-mail address: [email protected] (M. Fukuda). 0920-1211/$ — see front matter © 2007 Elsevier B.V. All rights reserved. doi:10.1016/j.eplepsyres.2007.06.007
16
J. Homma et al.
Introduction Hypothalamic hamartoma is a rare congenital disorder characterized by precocious puberty, seizures, and mental retardation. In patients with epilepsy, gelastic seizures usually develop during early childhood, followed later by development of other types of seizures, including tonic, complex partial, and generalized tonic clonic seizures (Berkovic et al., 1988). All of these seizures are medically intractable. The intrinsic epileptogenicity of hypothalamic hamartoma has been confirmed (Munari et al., 1995; Kuzniecky et al., 1997; Fukuda et al., 1999; Kahane et al., 2003), and surgical intervention is required for such lesions of the hypothalamus. Although various acceptable treatment options for hypothalamic hamartoma with intractable gelastic epilepsy have been reported, optimal strategies remain controversial in terms of efficacy and safety. In 1997, we began to use stereotactic thermocoagulation for treatment of relatively small hypothalamic hamartoma associated with gelastic seizures (Fukuda et al., 1999). Herein, we report the electrophysiologic characteristics of hypothalamic hamartomas causing gelastic seizures and the usefulness of stereotactic thermocoagulation surgery in five patients who underwent this treatment at our institution.
Subjects and methods Clinical presentation Five patients with hypothalamic hamartoma causing gelastic seizures were treated by stereotactic thermocoagulation at our cen-
Table 1
ter during the period October 1997 through February 2004 (Patient 1 was described in a previous report, Fukuda et al., 1999). Patient characteristics are summarized in Table 1. No patient had clinical signs of precocious puberty. Three patients (Patients 1, 2, and 3) showed mental deficiency preoperatively. In all cases, gelastic seizures began at 0—2 years of age, and various other types of seizures developed within a few years. All of these seizures were medically intractable. Time from the first seizure to surgery ranged from 8 to 36 years (20.2 ± 10.7 years). Pre- and postoperative evaluations included neurological examination, magnetic resonance imaging (MRI), long-term video-EEG monitoring with standard scalp electrodes, neuropsychological studies, and endocrinological evaluation by means of a triple stimulation test.
Chronic recording and advocative study Depth electrode recordings were obtained by means of a chronically implanted electrode in Patients 1, 2, and 4. Under local and intravenous propofol anesthesia, a quadripolar deep brain stimulation electrode (3387; Medtronic, Tokyo, Japan) was inserted into the hamartoma with the use of a Leksell Stereotactic System® (ELEKTA, Sweden). The depth electrode consisted of four contacts (with contact 0 being the most ventral contact and contact 3 being the most dorsal), which were 1.3 mm in diameter, 1.5 mm in length, and 1.5 mm apart. The target within the hamartoma was determined by computed tomography in Patient 1. In the other patients, MRI-based software (Leksell SurgiPlan® , ELEKTA) was used to determine the target (Fig. 1). In addition, several subdural strip electrodes were inserted bilaterally through burr holes and placed on the surface of the frontal cortices. Epileptic discharges detected by the implanted depth electrode in the hamartoma and subdural electrodes on the frontal cortex surface were monitored with simultaneous video recording over 7—21
Clinical characteristics of the five patients with hypothalamic hamartoma Patient 1
Patient 2
Patient 3
Patient 4
Patient 5
F
M
M
M
M
2 years
1 years
1 years 6 months
6 months
2 years
11 years 15 years
8 years 23 years
1 years 6 months 9 years
4 years 24 years
24 years
27 years
12 years
3 years 36 years 37 years 39 years
Follow-up period
8 years 6 months
4 years 3 months
3 years 1 months
2 years 2 months
Neuropsychology and behavior
Mild mental deficiency
Severe mental deficiency
Full-scale IQ
69 (WAIS-R)
Mild mental deficiency, aggressiveness 56 (WAIS-R)
2 years 10 months Normal social behavior
Endocrinology Types of seizure
No dysfunction GS + TS + CPS/GTC
No dysfunction GS + CPS, GS + GTC
MRI findings Hamartoma diameter Valdueza classification Delalande type
10 mm IIb II
15 mm IIb II
Sex Age At onset of gelastic seizures At onset of other seizures At 1st surgery At 2nd surgery At last follow-up
25 years
Normal social behavior
26 (Tanaka-Binet test) No dysfunction GS + TS, TS
102 (WAIS-R)
95 (WAIS-R)
No dysfunction GS + CPS
No dysfunction GS, GS + CPS
12 mm IIb II
10 mm IIb III
8 mm IIb III
GS: gelastic seizure; TS: tonic seizure; CPS: complex partial seizure; GTC: generalized tonic clonic seizure; IQ: intellectual quotient; WAIS-R: Wechsler Adult Intelligence Scale-Revised; Valdueza classification: anatomical and symptomatic classification proposed by Valdueza et al. (1994); Delalande type: anatomical classification proposed by Delalande and Fohlen (2003).
Stereotactic radiofrequency thermocoagulation
17
Fig. 1 (Left, center) Stereotactic plan produced by a Surgiplan workstation for Patient 5. Each circle represents a tentative lesion site, and the combination of lesions comprises a disconnecting plane at the boundary between the hypothalamus and the hamartoma. (Right) Coronal (upper) and axial (lower) T2-weighted magnetic resonance images obtained 1 week after the procedure reveal edematous change at the sites of thermocoagulation and trajectories of the coagulation needle.
days. In Patients 1 and 2, attempts were made to induce gelastic seizures by electrical stimulation of the depth electrode with a Neuropack (Nihon Kohden, Tokyo, Japan). Stimulation frequency was 50 Hz with a 5-s train and 0.2-ms pulse duration. Intensity was 0.2—12.0 mA.
Thermocoagulation Under local and intravenous propofol anesthesia, the Leksell stereotactic frame was attached to each patient. In Patients 1, 2, and 4 (first procedure), the depth electrode in the hamartoma was replaced by a coagulation needle 2 mm in diameter with a 4-mm uninsulated tip inserted to the same depth. In Patients 3 and 5 and in the second procedure for Patient 4, the target within the hamartoma was determined by SurgiPlan on the basis of preoperative MRI. After discontinuation of intravenous propofol anesthesia, test heating (60 ◦ C, 30 s) was performed to determine whether major complications were likely to occur. If patients did not experience any complication during the test, we then made one or several lesions (74 ◦ C, 60 s) according to the size of the hamartoma. In Patient 3, thermocoagulation was performed under general anesthesia because of his severe mental retardation.
Results General results In all cases, preoperative interictal scalp EEG showed sporadic generalized irregular slow-and-spike wave discharges with frontal predominance. During preoperative EEG monitoring, three or more seizures were recorded in Patients 1, 2, 4, and 5. In these patients, EEG showed generalized irregular rapid spike discharges followed by depression of background activity before ictal onset.
Results of preoperative hormonal examination with the triple stimulation test were within normal range in all patients, although Patient 1 showed slightly increased responses of luteinizing hormone, follicle-stimulating hormone, thyroid-stimulating hormone, and growth hormone. No patient showed clinical signs of precocious puberty. Preoperative MRI depicted the hamartoma as an isointense mass in all five patients. The tumors were 10, 15, 12, 10, and 8 mm in long-axis diameter. Each had sessile attachment to the hypothalamus and protruded toward the third ventricle and interpeduncular cistern, features corresponding to type IIb lesions described by Valdueza et al. (1994) (Fig. 2, upper row). The hamartoma was slightly deviated to the left in all but Patient 1. In two (Patients 1 and 4) of three patients in whom a depth electrode was placed in the hamartoma, we confirmed that the ictal discharge onset occurred on the contacts corresponding to the hamartoma itself followed by rapid spike discharges in the subdural electrodes placed on the cerebral cortex (Fig. 3). Electrical stimulation of the contacts of the depth electrode in the hamartoma gave rise to habitual gelastic seizures in all three patients (Fukuda et al., 1999). The postoperative follow-up period was 26—102 (50.6 ± 30.5) months. Three of the five patients (Patients 1, 4, and 5) were seizure-free at the time of the final follow-up examination (Table 2). In Patient 2, the gelastic seizures completely disappeared, but the other types of seizures remained despite a 90% decrease in frequency in comparison to the preoperative frequency. In only one patient (Patient 3), the gelastic seizures began to recur 3 months after surgery, although the seizure forms changed. Postoperatively, four patients suffered fever, two showed hyperphagia, and two showed character changes. However, these complications were transient. Results of hormonal
18
J. Homma et al.
Fig. 2 Pre- (upper row) and postoperative (lower row) coronal T1-weighted magnetic resonance images of the five patients. Each shows a relatively small hypothalamic hamartoma 10 mm or less in diameter with sessile attachment and protruding into the third ventricle and interpeduncular cistern.
Individual patients (Table 1)
hamartoma. Electrical stimulation of the contacts induced a gelastic seizure followed by tonic and generalized seizures. Creation of a lesion at the contact 1 site resulted in disappearance of the gelastic seizures immediately thereafter. The tonic seizures increased transiently but gradually disappeared within 1 month of the procedure. The patient is completely free of seizures 7 years later. Although no clear association with the procedure was identified, anxiety neurosis developed 1 year after the procedure, and the patient complained of visual hallucinations for 4 years after the surgery.
Patient 1 (previously reported in detail, Fukuda et al., 1999) The patient was a 15-year-old girl who had suffered from gelastic seizures beginning at 2 years of age. Prior to clinical gelastic seizures, ictal discharge onset occurred on the contacts of a depth electrode that had been inserted into the
Patient 2 The patient was a 23-year-old man who began experiencing gelastic seizures at age 1 year. After implantation of depth and subdural electrodes, no gelastic seizures occurred, and only two episodes of generalized seizures were noted. However, electrical stimulation of contact 1 of the depth
examination with the triple stimulation test performed 1 month postoperatively were within normal range in all patients and similar to the preoperative values. One year after surgery, MRI showed remarkable volume reduction of the hamartoma itself in Patients 4 and 5. In the other patients, attachment of the hamartoma had changed to various non-sessile forms, such as a constrictive (Patient 2), perforative (Patient 1), or degenerative (Patient 3) form (Fig. 2, lower row).
Fig. 3 Intracranial electroencephalogram obtained by means of a depth electrode inserted into the hamartoma and subdural grid electrodes placed on the front-parietal cortices via the same frontal burr hole. High frequency oscillation following a polyspike burst occurred in the hamartoma during the aura stage, and the patient felt a sudden impulse to laugh. No subsequent widespread spike-and-wave discharge was observed in the frontal cortex.
Anxiety neurosis Complications (transient)
Depth recording: recordings obtained by means of a chronically implanted depth electrode within hamartoma; GS: gelastic seizure; CPS: complex partial seizure.
Fever, character change Fever, hyperphagia, character change
Relief from all seizures Relief from all seizures
90% amelioration in frequency of gelastic seizures, GS disappeared Fever, hyperphagia Relief from all seizures
No change in frequency of gelastic seizures, CPS disappeared Fever
— 4/4 2/1(1st surgery) 2/2(2nd surgery)
— 2/2 + 4/2 + 1/1
+
Patient 3 Patient 1
Patient 2
Depth recording Number of coagulation lesions/tracks Outcome
Table 2
Results of stereotactic thermocoagulation
Patient 4
Patient 5
Stereotactic radiofrequency thermocoagulation
19 electrode within the hamartoma induced a habitual gelastic seizure. Four coagulation lesions were made at the center of the hamartoma, corresponding to the location of contact 1. Immediately after surgery, the gelastic seizures completely disappeared. However, tonic seizures continued to occur, with a 90% reduction in frequency in comparison to preoperative frequency. Postoperatively, the patient showed low-grade fever and hyperphagia for a few days. No postoperative decrease in the intelligence quotient was noted. Patient 3 The patient was an 11-year-old boy who showed abnormal laughter followed by transient symmetrical tonic posturing beginning at 1 year of age. He was severely mentally retarded (Tanaka-Binet intelligence score: 26). Coagulation lesions were created in the hamartoma at sites determined by MRI, and the patient was free from gelastic seizures for 2 months postoperatively. He showed low-grade fever for a few days after surgery. One month after surgery, he began experiencing frequent daytime tonic seizures, which lasted 4 weeks. Three months after surgery, the gelastic seizures recurred and increased gradually to the preoperative frequency. However, the patient did not lose consciousness during the gelastic seizures as he had preoperatively. Patient 4 The patient was a 37-year-old man who exhibited abnormal laughter beginning at 6 months of age. When the patient was 36 years old, an EEG recording obtained with depth and subdural electrodes showed no epileptic discharges during gelastic and complex partial seizures. During the aura stage, however, repetitive spikes were recorded at contacts 1, 2, and 3 of the depth electrode, and their amplitudes gradually increased (Fig. 3). Creation of coagulation lesions in the hamartoma, at sites corresponding to contacts 2 and 3, resulted in transient minor changes in the patient’s personality, manifesting as irascible demeanor lasting 1 week. Low-grade fever and hyperphagia were present for a few days after surgery. The seizures disappeared for 1 month but recurred at about 30% of the preoperative frequency. One year after the first surgery, a coagulation lesion was added to the narrow part between the hamartoma and hypothalamus, a site chosen on the basis of T1-weighted MRI with Leksell SurgiPlan® . Postoperatively, the frequency of all seizures gradually decreased, and seizures ceased completely 3 months later. No lasting complications including mental deterioration were seen after either procedure. Patient 5 The patient was a 24-year-old man who suffered laughing attacks that occurred during the early stage of his sleep beginning at 2 years of age. During the test heating and thermocoagulation, the patient showed generalized flushing and profuse sweating. Because the signs were transient and the other general conditions did not change, permanent coagulation lesions were made. The coagulation lesions within the hamartoma rendered him seizure-free immediately after surgery. He experienced mild fever for a few days and minor character changes such as marked cheerfulness and unusual sociability for about 1 week. No permanent complications occurred.
20
Discussion Stereotactic thermocoagulation for hypothalamic hamartoma resulted in freedom from gelastic seizures in four of our five patients by the time of the final examination. In three of these four patients, not only the gelastic seizures but also other types of seizures disappeared. Several symptoms, such as pyrexia, hyperphagia, and character change, occurred transiently after surgery, but no lasting complication including mental deterioration was seen. Thus, thermocoagulation for a relatively small hypothalamic hamartoma is a useful and safe means of controlling intractable gelastic seizures. In cases such as those described herein, the most common procedure at one time was direct surgical resection of the hamartoma by conventional craniotomy. However, complete resection was not always achieved, resulting in a hypothalamic disorder or mental decline due to injury of the mammillary bodies. In 1994, Valdueza et al. reported a relation between MRI discovery of sessile attachment of the hamartoma to the tuber cinereum or mammillary bodies and epileptic symptoms including gelastic seizures in 14 patients. In their study, an excellent or good outcome without morbidity was achieved in four patients, including three who underwent total removal of the hamartoma. Gelastic seizures remained unchanged in five patients, including four who underwent partial removal. Postoperatively, two patients had severe memory disturbance diagnosed as Korsakoff syndrome. Other investigators have recently described the usefulness of refined microsurgical techniques for resection of hypothalamic hamartoma to control gelastic seizures (Delalande and Fohlen, 2003; Fohlen et al., 2003; Harvey et al., 2003; Polkey, 2003; Choi et al., 2004). In particular, resection of the hypothalamic hamartoma via transcallosal approach provided relatively good seizure outcomes with only limited morbidity (Harvey et al., 2003; Polkey, 2003). A more recent study showed that 54% of patients were completely seizure-free and that 35% of patients had 90% improvement in total seizure frequency after transcallosal surgical resection (Ng et al., 2006). However, two patients (8%) had postoperative memory disturbance, and the other two (8%) had persisting endocrine disturbance requiring hormone replacement therapy. Stereotactic thermocoagulation for hypothalamic hamartoma has the advantage of greater safety than that previously reported for conventional surgery. In our cases, low-grade fever (four patients) and hyperphagia (two patients) occurred transiently as local hypothalamic symptoms. In two patients, transient minor character changes occurred due to the effect on the adjacent part of the limbic circuit in the hypothalamus. These transient signs can probably be attributed to perifocal edema occurring after thermocoagulation. Indeed, these symptoms resolved within 1 week after surgery, and no permanent complications were noted. This thermocoagulation technique has already been used for treatment of Parkinson’s disease, and its effectiveness and safety have been established. Four of our five patients (80%) were free of gelastic seizures by the time of the final examination, although one patient required two surgeries. The fifth patient had a recurrent gelastic seizure 3 months after surgery. Postoperative MRI revealed degeneration of the attachment of the
J. Homma et al. hamartoma, indicating accurate thermocoagulation. In this patient, epileptogenicity of the hypothalamic hamartoma could not be confirmed because he did not undergo chronic depth electrode monitoring due to severe mental retardation. Some remnant of the seizures may be attributed to an epileptogenic lesion existing in the extra-hypothalamic hamartoma or in the still-viable regions within the shrunken hamartoma. Three of four patients were free not only from gelastic seizures but also from other types of seizures, whereas the fourth patient continued to experience other types of seizures at decreased frequency. It is conceivable that other foci formed secondarily in regions outside the hypothalamus in this patient. Less-invasive treatment options for hypothalamic hamartoma have recently been developed, including endoscopeguided disconnection surgery (Delalande and Fohlen, 2003; Fohlen et al., 2003; Choi et al., 2004), and these provide relatively good outcomes with acceptable complication rates in comparison to those associated with conventional surgical approaches. Delalande and Fohlen (2003) described surgical disconnection of hamartomas by craniotomy or neuroendoscopy (Delalande and Fohlen, 2003; Fohlen et al., 2003). They proposed an anatomical classification based on the presence of a horizontal attachment on the inferior aspect of the hypothalamus, in which case disconnection is performed by craniotomy, or a vertical attachment in the third ventricle, in which case disconnection is performed by neuroendoscopy. They reported recently that frameless stereotactic endoscopic disconnection for hamartoma with ventricular localization (type II) provided excellent seizure outcomes. Ninety percent of patients affected by a type II hamartoma became seizure-free, and the remaining 10% improved (Procaccini et al., 2006). However, the authors recommended endoscopic disconnection followed by the pterional approach for the type III hamartoma (huge hamartoma characterized by partial intraventricular growth and horizontal or vertical plane of insertion). Consequently, patients with a type III hamartoma had worse outcomes than did those with a type II hamartoma. Stereotactic thermocoagulation can be applied to any type of hypothalamic hamartoma, regardless of the intraventricular location based on the plane of insertion on the hypothalamus (Delalande and Fohlen, 2003). From these findings, it appears that this technique offers some advantages over endoscopic disconnection surgery. It is probably important that coagulation lesions are produced not only in the center of the hamartoma but also in more rostral sites where fibers connecting to the outside of the hypothalamic hamartoma exist. Kahane et al. (2003) suggested that the mammillo-thalamo-cingulate tract serves as a discharge relay from the hypothalamic hamartoma toward the cortex; this theory is based on their depth recordings from the hypothalamic hamartoma and the mesial and lateral cortices. Indeed, our Patient 4 was seizure-free after a second surgery in which coagulation lesions were added to sites within the connection between the hamartoma and the floor of the third ventricle. Stereotactic radiosurgery with a gamma knife (Munari et al., 1995; Arita et al., 1998; Unger et al., 2000; Regis et al., 2000; Regis et al., 2004; Barajas et al., 2005) or interstitial radiosurgery with the use of stereotactically implanted 125 I-seeds (Schulze-Bonhage et al., 2004a,b) has also been
Stereotactic radiofrequency thermocoagulation used to treat epilepsy related to hypothalamic hamartoma. In a multicenter, retrospective study of gamma knife surgery (GKS) for hypothalamic hamartoma, four of eight patients became seizure-free, one experienced rare nocturnal seizures, one experienced rare partial seizures, and two showed some improvement, with a reduction in seizure frequency (Regis et al., 2000). In this patient series, no side effects were observed. However, seizure cessation occurred with a delay of at least 9 months. The longest period from GKS to seizure cessation was 36 months. Stereotactic thermocoagulation therapy has the advantage over GKS of earlier postoperative treatment effects. In our Patients 1, 2, and 5, gelastic seizures disappeared immediately after surgery. In Patient 4, the seizures did not disappear completely until 3 months after the second surgery. In addition, thermocoagulation therapy is probably safer than GKS. If local anesthesia is available, any side effects can be identified by test heating before permanent coagulation lesions are made. Additionally, radiation therapy always poses a risk of delayed complications such as radiation-induced necrosis and oncogenesis. Thus, radiation therapy may be inappropriate for treatment of hypothalamic hamartoma, which often occurs in young people. On the basis of our experience, we consider stereotactic thermocoagulation therapy the first-choice procedure for treatment of intractable gelastic epilepsy due to a small hypothalamic hamartoma. Long-term study of a large group of patients will allow us to confirm the usefulness and safety of this treatment.
References Arita, K., Kurisu, K., Iida, K., Hanaya, R., Akimitsu, T., Hibino, S., Pant, B., Hamasaki, M., Shinagawa, S., 1998. Subsidence of seizure induced by stereotactic radiation in a patient with hypothalamic hamartoma. Case report. J. Neurosurg. 89, 645—648. Barajas, M.A., Ramirez-Guzman, M.G., Rodriguez-Vazquez, C., Toledo-Buenrostro, V., Cuevas-Solorzano, A., RodriguezHernandez, G., 2005. Gamma knife surgery for hypothalamic hamartomas accompanied by medically intractable epilepsy and precocious puberty: experience in Mexico. J. Neurosurg. 102 (Suppl.), 53—55. Berkovic, S.F., Andermann, F., Melanson, D., Ethier, R.E., Feindel, W., Gloor, P., 1988. Hypothalamic hamartomas and ictal laughter: evolution of a characteristic epileptic syndrome and diagnostic value of magnetic resonance imaging. Ann. Neurol. 23, 429—439. Choi, J.U., Yang, K.H., Kim, T.G., Chang, J.H., Chang, J.W., Lee, B.I., Kim, D.S., 2004. Endoscopic disconnection for hypothalamic hamartoma with intractable seizure. Report of four cases. J. Neurosurg. 100 (5 Suppl. Pediatrics), 506—511. Delalande, O., Fohlen, M., 2003. Disconnecting surgical treatment of hypothalamic hamartoma in children and adults with refractory epilepsy and proposal of a new classification. Neurol. Med. Chir. (Tokyo) 43, 61—68.
21 Fohlen, M., Lellouch, A., Delalande, O., 2003. Hypothalamic hamartoma with refractory epilepsy: surgical procedures and results in 18 patients. Epileptic Disord. 5, 267—273. Fukuda, M., Kameyama, S., Wachi, M., Tanaka, R., 1999. Stereotaxy for hypothalamic hamartoma with intractable gelastic seizures: technical case report. Neurosurgery 44, 1347—1350. Harvey, A.S., Freeman, J.L., Berkovic, S.F., Rosenfeld, J.V., 2003. Transcallosal resection of hypothalamic hamartomas in patients with intractable epilepsy. Epileptic Disord. 5, 257—265. Kahane, P., Ryvlin, P., Hoffmann, D., Minotti, L., Benabid, A.L., 2003. From hypothalamic hamartoma to cortex: what can be learnt from depth recordings and stimulation? Epileptic Disord. 5, 205—217. Kuzniecky, R., Guthrie, B., Mountz, J., Bebin, M., Faught, E., Gilliam, F., Liu, H.G., 1997. Intrinsic epileptogenesis of hypothalamic hamartomas in gelastic epilepsy. Ann. Neurol. 42, 60— 67. Munari, C., Kahane, P., Francione, S., Hoffmann, D., Tassi, L., Cusmai, R., Vigevano, F., Pasquier, B., Betti, O.O., 1995. Role of the hypothalamic hamartoma in the genesis of gelastic fits (a videostereo-EEG study). Electroencephalogr. Clin. Neurophysiol. 95, 154—160. Ng, Y., Rekate, H.L., Prenger, E.C., Chung, S.S., Feiz-Erfan, I., Wang, N.C., Verland, M.R., Kerrigan, J.F., 2006. Transcallosal resection of hypothalamic hamartoma for intractable epilepsy. Epilepsia 47, 1192—1202. Polkey, C.E., 2003. Resective surgery for hypothalamic hamartoma. Epileptic Disord. 5, 281—286. Procaccini, E., Dorfmuller, G., Fohlen, M., Bulteau, C., Delalande, O., 2006. Surgical management of hypothalamic hamartomas with epilepsy: the stereoendoscopic approach. Neurosurgery 59 (ONS Suppl. 4) (ONS-336-ONS-346). Regis, J., Bartolomei, F., de Toffol, B., Genton, P., Kobayashi, T., Mori, Y., Takakura, K., Hori, T., Inoue, H., Schrottner, O., Pendel, G., Wolf, A., Arita, K., Chauvel, P., 2000. Gamma knife surgery for epilepsy related to hypothalamic hamartomas. Neurosurgery 47, 1343—1351. Regis, J., Hayashi, M., Eupierre, L.P., Villeneuve, N., Bartolomei, F., Brue, T., Chauvel, P., 2004. Gamma knife surgery for epilepsy related to hypothalamic hamartoma. Acta Neurochir. Suppl. 91, 33—50. Schulze-Bonhage, A., Homberg, V., Trippel, M., Keimer, R., Elger, C.E., Warnke, P.C., Ostertag, C., 2004a. Interstitial radiosurgery in the treatment of gelastic epilepsy due to hypothalamic hamartomas. Neurology 62, 644—647. Schulze-Bonhage, A., Quiske, A., Homberg, V., Trippel, M., Wagner, K., Frings, L., Bast, T., Huppertz, H.J., Warnke, C., Ostertag, Ch., 2004b. Effect of interstitial stereotactic radiosurgery on behavior and subjective handicap of epilepsy in patients with gelastic epilepsy. Epilepsy Behav. 5, 94—101. Unger, F., Schrottner, O., Haselsberger, K., Korner, E., Ploier, R., Pendl, G., 2000. Gamma knife radiosurgery for hypothalamic hamartomas in patients with medically intractable epilepsy and precocious puberty. Report of two cases. J. Neurosurg. 92, 726—731. Valdueza, J.M., Cristante, L., Dammann, O., Bentele, K., Vortmeyer, A., Saeger, W., Padberg, B., Freitag, J., Herrmann, H.D., 1994. Hypothalamic hamartomas with special reference to gelastic epilepsy and surgery. Neurosurgery 34, 949—958.
journal homepage: www.elsevier.com/locate/epilepsyres
CLINICAL RESEARCH
Stereotactic radiofrequency thermocoagulation for hypothalamic hamartoma with intractable gelastic seizures Junpei Homma a, Shigeki Kameyama a, Hiroshi Masuda a, Takehiko Ueno a, Ayataka Fujimoto a, Makoto Oishi a,b, Masafumi Fukuda a,b,∗ a
Department of Neurosurgery, Epilepsy Center, Nishi-Niigata Chuo National Hospital, 1-14-1 Masago, Niigata 950-2085, Japan Department of Neurosurgery, Brain Research Institute, University of Niigata, 1-757 Asahimachi-dori, Niigata-City 951-8585, Japan
b
Received 24 December 2006; received in revised form 19 May 2007; accepted 3 June 2007 Available online 23 July 2007
KEYWORDS Gelastic epilepsy; Hypothalamic hamartoma; Radiofrequency thermocoagulation; Stereotactic surgery; Surgical treatment
Summary Management of hypothalamic hamartoma with intractable gelastic epilepsy remains controversial. We have used stereotactic thermocoagulation for treatment of hypothalamic hamartoma with intractable gelastic epilepsy since 1997. Herein, we review our experience in five cases to clarify the usefulness of this treatment. A total of five patients with hypothalamic hamartoma were treated by stereotactic thermocoagulation at our hospital during the period October 1997 through February 2004. In all patients, the hamartoma was less than 10 mm in diameter and was located on the floor of the third ventricle with sessile attachment to the wall. To identify ictal onset, chronic intracranial electroencephalography was performed in three patients with the use of a depth electrode implanted in the hamartoma. Attempts were made to induce gelastic seizure by electrical stimulation of the hamartoma in three patients. After magnetic resonance imaging-guided targeting, radiofrequency thermocoagulation of the boundary between the hamartoma and normal hypothalamus was performed to achieve disconnection effects. Marked reductions in seizure frequency were obtained in all cases, with three patients becoming seizure-free after the procedure. No intraoperative complications occurred except in one patient who experienced acute and transient panidrosis with hot flushes during coagulation. Our results suggest that stereotactic thermocoagulation of hypothalamic hamartoma is an acceptable treatment option for patients with intractable gelastic seizures. © 2007 Elsevier B.V. All rights reserved.
∗
Corresponding author at: Department of Neurosurgery, Brain Research Institute, 1-757 Asahimachi-dori, Niigata-City 951-8585, Japan. Tel.: +81 25 227 0652; fax: +81 25 227 0819. E-mail address: [email protected] (M. Fukuda). 0920-1211/$ — see front matter © 2007 Elsevier B.V. All rights reserved. doi:10.1016/j.eplepsyres.2007.06.007
16
J. Homma et al.
Introduction Hypothalamic hamartoma is a rare congenital disorder characterized by precocious puberty, seizures, and mental retardation. In patients with epilepsy, gelastic seizures usually develop during early childhood, followed later by development of other types of seizures, including tonic, complex partial, and generalized tonic clonic seizures (Berkovic et al., 1988). All of these seizures are medically intractable. The intrinsic epileptogenicity of hypothalamic hamartoma has been confirmed (Munari et al., 1995; Kuzniecky et al., 1997; Fukuda et al., 1999; Kahane et al., 2003), and surgical intervention is required for such lesions of the hypothalamus. Although various acceptable treatment options for hypothalamic hamartoma with intractable gelastic epilepsy have been reported, optimal strategies remain controversial in terms of efficacy and safety. In 1997, we began to use stereotactic thermocoagulation for treatment of relatively small hypothalamic hamartoma associated with gelastic seizures (Fukuda et al., 1999). Herein, we report the electrophysiologic characteristics of hypothalamic hamartomas causing gelastic seizures and the usefulness of stereotactic thermocoagulation surgery in five patients who underwent this treatment at our institution.
Subjects and methods Clinical presentation Five patients with hypothalamic hamartoma causing gelastic seizures were treated by stereotactic thermocoagulation at our cen-
Table 1
ter during the period October 1997 through February 2004 (Patient 1 was described in a previous report, Fukuda et al., 1999). Patient characteristics are summarized in Table 1. No patient had clinical signs of precocious puberty. Three patients (Patients 1, 2, and 3) showed mental deficiency preoperatively. In all cases, gelastic seizures began at 0—2 years of age, and various other types of seizures developed within a few years. All of these seizures were medically intractable. Time from the first seizure to surgery ranged from 8 to 36 years (20.2 ± 10.7 years). Pre- and postoperative evaluations included neurological examination, magnetic resonance imaging (MRI), long-term video-EEG monitoring with standard scalp electrodes, neuropsychological studies, and endocrinological evaluation by means of a triple stimulation test.
Chronic recording and advocative study Depth electrode recordings were obtained by means of a chronically implanted electrode in Patients 1, 2, and 4. Under local and intravenous propofol anesthesia, a quadripolar deep brain stimulation electrode (3387; Medtronic, Tokyo, Japan) was inserted into the hamartoma with the use of a Leksell Stereotactic System® (ELEKTA, Sweden). The depth electrode consisted of four contacts (with contact 0 being the most ventral contact and contact 3 being the most dorsal), which were 1.3 mm in diameter, 1.5 mm in length, and 1.5 mm apart. The target within the hamartoma was determined by computed tomography in Patient 1. In the other patients, MRI-based software (Leksell SurgiPlan® , ELEKTA) was used to determine the target (Fig. 1). In addition, several subdural strip electrodes were inserted bilaterally through burr holes and placed on the surface of the frontal cortices. Epileptic discharges detected by the implanted depth electrode in the hamartoma and subdural electrodes on the frontal cortex surface were monitored with simultaneous video recording over 7—21
Clinical characteristics of the five patients with hypothalamic hamartoma Patient 1
Patient 2
Patient 3
Patient 4
Patient 5
F
M
M
M
M
2 years
1 years
1 years 6 months
6 months
2 years
11 years 15 years
8 years 23 years
1 years 6 months 9 years
4 years 24 years
24 years
27 years
12 years
3 years 36 years 37 years 39 years
Follow-up period
8 years 6 months
4 years 3 months
3 years 1 months
2 years 2 months
Neuropsychology and behavior
Mild mental deficiency
Severe mental deficiency
Full-scale IQ
69 (WAIS-R)
Mild mental deficiency, aggressiveness 56 (WAIS-R)
2 years 10 months Normal social behavior
Endocrinology Types of seizure
No dysfunction GS + TS + CPS/GTC
No dysfunction GS + CPS, GS + GTC
MRI findings Hamartoma diameter Valdueza classification Delalande type
10 mm IIb II
15 mm IIb II
Sex Age At onset of gelastic seizures At onset of other seizures At 1st surgery At 2nd surgery At last follow-up
25 years
Normal social behavior
26 (Tanaka-Binet test) No dysfunction GS + TS, TS
102 (WAIS-R)
95 (WAIS-R)
No dysfunction GS + CPS
No dysfunction GS, GS + CPS
12 mm IIb II
10 mm IIb III
8 mm IIb III
GS: gelastic seizure; TS: tonic seizure; CPS: complex partial seizure; GTC: generalized tonic clonic seizure; IQ: intellectual quotient; WAIS-R: Wechsler Adult Intelligence Scale-Revised; Valdueza classification: anatomical and symptomatic classification proposed by Valdueza et al. (1994); Delalande type: anatomical classification proposed by Delalande and Fohlen (2003).
Stereotactic radiofrequency thermocoagulation
17
Fig. 1 (Left, center) Stereotactic plan produced by a Surgiplan workstation for Patient 5. Each circle represents a tentative lesion site, and the combination of lesions comprises a disconnecting plane at the boundary between the hypothalamus and the hamartoma. (Right) Coronal (upper) and axial (lower) T2-weighted magnetic resonance images obtained 1 week after the procedure reveal edematous change at the sites of thermocoagulation and trajectories of the coagulation needle.
days. In Patients 1 and 2, attempts were made to induce gelastic seizures by electrical stimulation of the depth electrode with a Neuropack (Nihon Kohden, Tokyo, Japan). Stimulation frequency was 50 Hz with a 5-s train and 0.2-ms pulse duration. Intensity was 0.2—12.0 mA.
Thermocoagulation Under local and intravenous propofol anesthesia, the Leksell stereotactic frame was attached to each patient. In Patients 1, 2, and 4 (first procedure), the depth electrode in the hamartoma was replaced by a coagulation needle 2 mm in diameter with a 4-mm uninsulated tip inserted to the same depth. In Patients 3 and 5 and in the second procedure for Patient 4, the target within the hamartoma was determined by SurgiPlan on the basis of preoperative MRI. After discontinuation of intravenous propofol anesthesia, test heating (60 ◦ C, 30 s) was performed to determine whether major complications were likely to occur. If patients did not experience any complication during the test, we then made one or several lesions (74 ◦ C, 60 s) according to the size of the hamartoma. In Patient 3, thermocoagulation was performed under general anesthesia because of his severe mental retardation.
Results General results In all cases, preoperative interictal scalp EEG showed sporadic generalized irregular slow-and-spike wave discharges with frontal predominance. During preoperative EEG monitoring, three or more seizures were recorded in Patients 1, 2, 4, and 5. In these patients, EEG showed generalized irregular rapid spike discharges followed by depression of background activity before ictal onset.
Results of preoperative hormonal examination with the triple stimulation test were within normal range in all patients, although Patient 1 showed slightly increased responses of luteinizing hormone, follicle-stimulating hormone, thyroid-stimulating hormone, and growth hormone. No patient showed clinical signs of precocious puberty. Preoperative MRI depicted the hamartoma as an isointense mass in all five patients. The tumors were 10, 15, 12, 10, and 8 mm in long-axis diameter. Each had sessile attachment to the hypothalamus and protruded toward the third ventricle and interpeduncular cistern, features corresponding to type IIb lesions described by Valdueza et al. (1994) (Fig. 2, upper row). The hamartoma was slightly deviated to the left in all but Patient 1. In two (Patients 1 and 4) of three patients in whom a depth electrode was placed in the hamartoma, we confirmed that the ictal discharge onset occurred on the contacts corresponding to the hamartoma itself followed by rapid spike discharges in the subdural electrodes placed on the cerebral cortex (Fig. 3). Electrical stimulation of the contacts of the depth electrode in the hamartoma gave rise to habitual gelastic seizures in all three patients (Fukuda et al., 1999). The postoperative follow-up period was 26—102 (50.6 ± 30.5) months. Three of the five patients (Patients 1, 4, and 5) were seizure-free at the time of the final follow-up examination (Table 2). In Patient 2, the gelastic seizures completely disappeared, but the other types of seizures remained despite a 90% decrease in frequency in comparison to the preoperative frequency. In only one patient (Patient 3), the gelastic seizures began to recur 3 months after surgery, although the seizure forms changed. Postoperatively, four patients suffered fever, two showed hyperphagia, and two showed character changes. However, these complications were transient. Results of hormonal
18
J. Homma et al.
Fig. 2 Pre- (upper row) and postoperative (lower row) coronal T1-weighted magnetic resonance images of the five patients. Each shows a relatively small hypothalamic hamartoma 10 mm or less in diameter with sessile attachment and protruding into the third ventricle and interpeduncular cistern.
Individual patients (Table 1)
hamartoma. Electrical stimulation of the contacts induced a gelastic seizure followed by tonic and generalized seizures. Creation of a lesion at the contact 1 site resulted in disappearance of the gelastic seizures immediately thereafter. The tonic seizures increased transiently but gradually disappeared within 1 month of the procedure. The patient is completely free of seizures 7 years later. Although no clear association with the procedure was identified, anxiety neurosis developed 1 year after the procedure, and the patient complained of visual hallucinations for 4 years after the surgery.
Patient 1 (previously reported in detail, Fukuda et al., 1999) The patient was a 15-year-old girl who had suffered from gelastic seizures beginning at 2 years of age. Prior to clinical gelastic seizures, ictal discharge onset occurred on the contacts of a depth electrode that had been inserted into the
Patient 2 The patient was a 23-year-old man who began experiencing gelastic seizures at age 1 year. After implantation of depth and subdural electrodes, no gelastic seizures occurred, and only two episodes of generalized seizures were noted. However, electrical stimulation of contact 1 of the depth
examination with the triple stimulation test performed 1 month postoperatively were within normal range in all patients and similar to the preoperative values. One year after surgery, MRI showed remarkable volume reduction of the hamartoma itself in Patients 4 and 5. In the other patients, attachment of the hamartoma had changed to various non-sessile forms, such as a constrictive (Patient 2), perforative (Patient 1), or degenerative (Patient 3) form (Fig. 2, lower row).
Fig. 3 Intracranial electroencephalogram obtained by means of a depth electrode inserted into the hamartoma and subdural grid electrodes placed on the front-parietal cortices via the same frontal burr hole. High frequency oscillation following a polyspike burst occurred in the hamartoma during the aura stage, and the patient felt a sudden impulse to laugh. No subsequent widespread spike-and-wave discharge was observed in the frontal cortex.
Anxiety neurosis Complications (transient)
Depth recording: recordings obtained by means of a chronically implanted depth electrode within hamartoma; GS: gelastic seizure; CPS: complex partial seizure.
Fever, character change Fever, hyperphagia, character change
Relief from all seizures Relief from all seizures
90% amelioration in frequency of gelastic seizures, GS disappeared Fever, hyperphagia Relief from all seizures
No change in frequency of gelastic seizures, CPS disappeared Fever
— 4/4 2/1(1st surgery) 2/2(2nd surgery)
— 2/2 + 4/2 + 1/1
+
Patient 3 Patient 1
Patient 2
Depth recording Number of coagulation lesions/tracks Outcome
Table 2
Results of stereotactic thermocoagulation
Patient 4
Patient 5
Stereotactic radiofrequency thermocoagulation
19 electrode within the hamartoma induced a habitual gelastic seizure. Four coagulation lesions were made at the center of the hamartoma, corresponding to the location of contact 1. Immediately after surgery, the gelastic seizures completely disappeared. However, tonic seizures continued to occur, with a 90% reduction in frequency in comparison to preoperative frequency. Postoperatively, the patient showed low-grade fever and hyperphagia for a few days. No postoperative decrease in the intelligence quotient was noted. Patient 3 The patient was an 11-year-old boy who showed abnormal laughter followed by transient symmetrical tonic posturing beginning at 1 year of age. He was severely mentally retarded (Tanaka-Binet intelligence score: 26). Coagulation lesions were created in the hamartoma at sites determined by MRI, and the patient was free from gelastic seizures for 2 months postoperatively. He showed low-grade fever for a few days after surgery. One month after surgery, he began experiencing frequent daytime tonic seizures, which lasted 4 weeks. Three months after surgery, the gelastic seizures recurred and increased gradually to the preoperative frequency. However, the patient did not lose consciousness during the gelastic seizures as he had preoperatively. Patient 4 The patient was a 37-year-old man who exhibited abnormal laughter beginning at 6 months of age. When the patient was 36 years old, an EEG recording obtained with depth and subdural electrodes showed no epileptic discharges during gelastic and complex partial seizures. During the aura stage, however, repetitive spikes were recorded at contacts 1, 2, and 3 of the depth electrode, and their amplitudes gradually increased (Fig. 3). Creation of coagulation lesions in the hamartoma, at sites corresponding to contacts 2 and 3, resulted in transient minor changes in the patient’s personality, manifesting as irascible demeanor lasting 1 week. Low-grade fever and hyperphagia were present for a few days after surgery. The seizures disappeared for 1 month but recurred at about 30% of the preoperative frequency. One year after the first surgery, a coagulation lesion was added to the narrow part between the hamartoma and hypothalamus, a site chosen on the basis of T1-weighted MRI with Leksell SurgiPlan® . Postoperatively, the frequency of all seizures gradually decreased, and seizures ceased completely 3 months later. No lasting complications including mental deterioration were seen after either procedure. Patient 5 The patient was a 24-year-old man who suffered laughing attacks that occurred during the early stage of his sleep beginning at 2 years of age. During the test heating and thermocoagulation, the patient showed generalized flushing and profuse sweating. Because the signs were transient and the other general conditions did not change, permanent coagulation lesions were made. The coagulation lesions within the hamartoma rendered him seizure-free immediately after surgery. He experienced mild fever for a few days and minor character changes such as marked cheerfulness and unusual sociability for about 1 week. No permanent complications occurred.
20
Discussion Stereotactic thermocoagulation for hypothalamic hamartoma resulted in freedom from gelastic seizures in four of our five patients by the time of the final examination. In three of these four patients, not only the gelastic seizures but also other types of seizures disappeared. Several symptoms, such as pyrexia, hyperphagia, and character change, occurred transiently after surgery, but no lasting complication including mental deterioration was seen. Thus, thermocoagulation for a relatively small hypothalamic hamartoma is a useful and safe means of controlling intractable gelastic seizures. In cases such as those described herein, the most common procedure at one time was direct surgical resection of the hamartoma by conventional craniotomy. However, complete resection was not always achieved, resulting in a hypothalamic disorder or mental decline due to injury of the mammillary bodies. In 1994, Valdueza et al. reported a relation between MRI discovery of sessile attachment of the hamartoma to the tuber cinereum or mammillary bodies and epileptic symptoms including gelastic seizures in 14 patients. In their study, an excellent or good outcome without morbidity was achieved in four patients, including three who underwent total removal of the hamartoma. Gelastic seizures remained unchanged in five patients, including four who underwent partial removal. Postoperatively, two patients had severe memory disturbance diagnosed as Korsakoff syndrome. Other investigators have recently described the usefulness of refined microsurgical techniques for resection of hypothalamic hamartoma to control gelastic seizures (Delalande and Fohlen, 2003; Fohlen et al., 2003; Harvey et al., 2003; Polkey, 2003; Choi et al., 2004). In particular, resection of the hypothalamic hamartoma via transcallosal approach provided relatively good seizure outcomes with only limited morbidity (Harvey et al., 2003; Polkey, 2003). A more recent study showed that 54% of patients were completely seizure-free and that 35% of patients had 90% improvement in total seizure frequency after transcallosal surgical resection (Ng et al., 2006). However, two patients (8%) had postoperative memory disturbance, and the other two (8%) had persisting endocrine disturbance requiring hormone replacement therapy. Stereotactic thermocoagulation for hypothalamic hamartoma has the advantage of greater safety than that previously reported for conventional surgery. In our cases, low-grade fever (four patients) and hyperphagia (two patients) occurred transiently as local hypothalamic symptoms. In two patients, transient minor character changes occurred due to the effect on the adjacent part of the limbic circuit in the hypothalamus. These transient signs can probably be attributed to perifocal edema occurring after thermocoagulation. Indeed, these symptoms resolved within 1 week after surgery, and no permanent complications were noted. This thermocoagulation technique has already been used for treatment of Parkinson’s disease, and its effectiveness and safety have been established. Four of our five patients (80%) were free of gelastic seizures by the time of the final examination, although one patient required two surgeries. The fifth patient had a recurrent gelastic seizure 3 months after surgery. Postoperative MRI revealed degeneration of the attachment of the
J. Homma et al. hamartoma, indicating accurate thermocoagulation. In this patient, epileptogenicity of the hypothalamic hamartoma could not be confirmed because he did not undergo chronic depth electrode monitoring due to severe mental retardation. Some remnant of the seizures may be attributed to an epileptogenic lesion existing in the extra-hypothalamic hamartoma or in the still-viable regions within the shrunken hamartoma. Three of four patients were free not only from gelastic seizures but also from other types of seizures, whereas the fourth patient continued to experience other types of seizures at decreased frequency. It is conceivable that other foci formed secondarily in regions outside the hypothalamus in this patient. Less-invasive treatment options for hypothalamic hamartoma have recently been developed, including endoscopeguided disconnection surgery (Delalande and Fohlen, 2003; Fohlen et al., 2003; Choi et al., 2004), and these provide relatively good outcomes with acceptable complication rates in comparison to those associated with conventional surgical approaches. Delalande and Fohlen (2003) described surgical disconnection of hamartomas by craniotomy or neuroendoscopy (Delalande and Fohlen, 2003; Fohlen et al., 2003). They proposed an anatomical classification based on the presence of a horizontal attachment on the inferior aspect of the hypothalamus, in which case disconnection is performed by craniotomy, or a vertical attachment in the third ventricle, in which case disconnection is performed by neuroendoscopy. They reported recently that frameless stereotactic endoscopic disconnection for hamartoma with ventricular localization (type II) provided excellent seizure outcomes. Ninety percent of patients affected by a type II hamartoma became seizure-free, and the remaining 10% improved (Procaccini et al., 2006). However, the authors recommended endoscopic disconnection followed by the pterional approach for the type III hamartoma (huge hamartoma characterized by partial intraventricular growth and horizontal or vertical plane of insertion). Consequently, patients with a type III hamartoma had worse outcomes than did those with a type II hamartoma. Stereotactic thermocoagulation can be applied to any type of hypothalamic hamartoma, regardless of the intraventricular location based on the plane of insertion on the hypothalamus (Delalande and Fohlen, 2003). From these findings, it appears that this technique offers some advantages over endoscopic disconnection surgery. It is probably important that coagulation lesions are produced not only in the center of the hamartoma but also in more rostral sites where fibers connecting to the outside of the hypothalamic hamartoma exist. Kahane et al. (2003) suggested that the mammillo-thalamo-cingulate tract serves as a discharge relay from the hypothalamic hamartoma toward the cortex; this theory is based on their depth recordings from the hypothalamic hamartoma and the mesial and lateral cortices. Indeed, our Patient 4 was seizure-free after a second surgery in which coagulation lesions were added to sites within the connection between the hamartoma and the floor of the third ventricle. Stereotactic radiosurgery with a gamma knife (Munari et al., 1995; Arita et al., 1998; Unger et al., 2000; Regis et al., 2000; Regis et al., 2004; Barajas et al., 2005) or interstitial radiosurgery with the use of stereotactically implanted 125 I-seeds (Schulze-Bonhage et al., 2004a,b) has also been
Stereotactic radiofrequency thermocoagulation used to treat epilepsy related to hypothalamic hamartoma. In a multicenter, retrospective study of gamma knife surgery (GKS) for hypothalamic hamartoma, four of eight patients became seizure-free, one experienced rare nocturnal seizures, one experienced rare partial seizures, and two showed some improvement, with a reduction in seizure frequency (Regis et al., 2000). In this patient series, no side effects were observed. However, seizure cessation occurred with a delay of at least 9 months. The longest period from GKS to seizure cessation was 36 months. Stereotactic thermocoagulation therapy has the advantage over GKS of earlier postoperative treatment effects. In our Patients 1, 2, and 5, gelastic seizures disappeared immediately after surgery. In Patient 4, the seizures did not disappear completely until 3 months after the second surgery. In addition, thermocoagulation therapy is probably safer than GKS. If local anesthesia is available, any side effects can be identified by test heating before permanent coagulation lesions are made. Additionally, radiation therapy always poses a risk of delayed complications such as radiation-induced necrosis and oncogenesis. Thus, radiation therapy may be inappropriate for treatment of hypothalamic hamartoma, which often occurs in young people. On the basis of our experience, we consider stereotactic thermocoagulation therapy the first-choice procedure for treatment of intractable gelastic epilepsy due to a small hypothalamic hamartoma. Long-term study of a large group of patients will allow us to confirm the usefulness and safety of this treatment.
References Arita, K., Kurisu, K., Iida, K., Hanaya, R., Akimitsu, T., Hibino, S., Pant, B., Hamasaki, M., Shinagawa, S., 1998. Subsidence of seizure induced by stereotactic radiation in a patient with hypothalamic hamartoma. Case report. J. Neurosurg. 89, 645—648. Barajas, M.A., Ramirez-Guzman, M.G., Rodriguez-Vazquez, C., Toledo-Buenrostro, V., Cuevas-Solorzano, A., RodriguezHernandez, G., 2005. Gamma knife surgery for hypothalamic hamartomas accompanied by medically intractable epilepsy and precocious puberty: experience in Mexico. J. Neurosurg. 102 (Suppl.), 53—55. Berkovic, S.F., Andermann, F., Melanson, D., Ethier, R.E., Feindel, W., Gloor, P., 1988. Hypothalamic hamartomas and ictal laughter: evolution of a characteristic epileptic syndrome and diagnostic value of magnetic resonance imaging. Ann. Neurol. 23, 429—439. Choi, J.U., Yang, K.H., Kim, T.G., Chang, J.H., Chang, J.W., Lee, B.I., Kim, D.S., 2004. Endoscopic disconnection for hypothalamic hamartoma with intractable seizure. Report of four cases. J. Neurosurg. 100 (5 Suppl. Pediatrics), 506—511. Delalande, O., Fohlen, M., 2003. Disconnecting surgical treatment of hypothalamic hamartoma in children and adults with refractory epilepsy and proposal of a new classification. Neurol. Med. Chir. (Tokyo) 43, 61—68.
21 Fohlen, M., Lellouch, A., Delalande, O., 2003. Hypothalamic hamartoma with refractory epilepsy: surgical procedures and results in 18 patients. Epileptic Disord. 5, 267—273. Fukuda, M., Kameyama, S., Wachi, M., Tanaka, R., 1999. Stereotaxy for hypothalamic hamartoma with intractable gelastic seizures: technical case report. Neurosurgery 44, 1347—1350. Harvey, A.S., Freeman, J.L., Berkovic, S.F., Rosenfeld, J.V., 2003. Transcallosal resection of hypothalamic hamartomas in patients with intractable epilepsy. Epileptic Disord. 5, 257—265. Kahane, P., Ryvlin, P., Hoffmann, D., Minotti, L., Benabid, A.L., 2003. From hypothalamic hamartoma to cortex: what can be learnt from depth recordings and stimulation? Epileptic Disord. 5, 205—217. Kuzniecky, R., Guthrie, B., Mountz, J., Bebin, M., Faught, E., Gilliam, F., Liu, H.G., 1997. Intrinsic epileptogenesis of hypothalamic hamartomas in gelastic epilepsy. Ann. Neurol. 42, 60— 67. Munari, C., Kahane, P., Francione, S., Hoffmann, D., Tassi, L., Cusmai, R., Vigevano, F., Pasquier, B., Betti, O.O., 1995. Role of the hypothalamic hamartoma in the genesis of gelastic fits (a videostereo-EEG study). Electroencephalogr. Clin. Neurophysiol. 95, 154—160. Ng, Y., Rekate, H.L., Prenger, E.C., Chung, S.S., Feiz-Erfan, I., Wang, N.C., Verland, M.R., Kerrigan, J.F., 2006. Transcallosal resection of hypothalamic hamartoma for intractable epilepsy. Epilepsia 47, 1192—1202. Polkey, C.E., 2003. Resective surgery for hypothalamic hamartoma. Epileptic Disord. 5, 281—286. Procaccini, E., Dorfmuller, G., Fohlen, M., Bulteau, C., Delalande, O., 2006. Surgical management of hypothalamic hamartomas with epilepsy: the stereoendoscopic approach. Neurosurgery 59 (ONS Suppl. 4) (ONS-336-ONS-346). Regis, J., Bartolomei, F., de Toffol, B., Genton, P., Kobayashi, T., Mori, Y., Takakura, K., Hori, T., Inoue, H., Schrottner, O., Pendel, G., Wolf, A., Arita, K., Chauvel, P., 2000. Gamma knife surgery for epilepsy related to hypothalamic hamartomas. Neurosurgery 47, 1343—1351. Regis, J., Hayashi, M., Eupierre, L.P., Villeneuve, N., Bartolomei, F., Brue, T., Chauvel, P., 2004. Gamma knife surgery for epilepsy related to hypothalamic hamartoma. Acta Neurochir. Suppl. 91, 33—50. Schulze-Bonhage, A., Homberg, V., Trippel, M., Keimer, R., Elger, C.E., Warnke, P.C., Ostertag, C., 2004a. Interstitial radiosurgery in the treatment of gelastic epilepsy due to hypothalamic hamartomas. Neurology 62, 644—647. Schulze-Bonhage, A., Quiske, A., Homberg, V., Trippel, M., Wagner, K., Frings, L., Bast, T., Huppertz, H.J., Warnke, C., Ostertag, Ch., 2004b. Effect of interstitial stereotactic radiosurgery on behavior and subjective handicap of epilepsy in patients with gelastic epilepsy. Epilepsy Behav. 5, 94—101. Unger, F., Schrottner, O., Haselsberger, K., Korner, E., Ploier, R., Pendl, G., 2000. Gamma knife radiosurgery for hypothalamic hamartomas in patients with medically intractable epilepsy and precocious puberty. Report of two cases. J. Neurosurg. 92, 726—731. Valdueza, J.M., Cristante, L., Dammann, O., Bentele, K., Vortmeyer, A., Saeger, W., Padberg, B., Freitag, J., Herrmann, H.D., 1994. Hypothalamic hamartomas with special reference to gelastic epilepsy and surgery. Neurosurgery 34, 949—958.