Gamma Knife Surgery for EpilepsyRelated to Hypothalamic HamartomasJean Régis, MD,* Didier Scavarda, MD,‡ Manabu Tamura, MD,* Nathalie Villeneuve, MD,†

Fabrice Bartolomei, MD, PhD,† Thierry Brue, MD, PhD,‡,§ Isabelle Morange, MD,§

David Dafonseca, MD,& and Patrick Chauvel, MD†

Numerous neurosurgical approaches are available for children presenting with hypotha-lamic hamartomas (HHs) associated with severe epilepsy. A concern regarding the impair-ment of short-term memory after resective surgery is promoting the exploration of lessinvasive alternatives like radiosurgery. Gamma knife radiosurgery (GKS) can lead to a realreversal of the epileptic encephalopathy. Three years after radiosurgery, 60% of thechildren have an excellent result with complete seizure cessation in 40% and rare nondis-abling seizures in 20%, often in association with dramatic behavioral and cognitive im-provement. No permanent neurologic complications have thus far been reported. Raretransient cases of poikilothermia have been observed. GKS is clearly the safer approach forthese difficult patients. Young patients with severe epilepsy and neurocognitive comorbid-ity must be treated by using a curative approach as early as possible. Topological type(according to our original classification) is the major feature for selection of the besttreatment strategy. Type I HH deeply embedded in the hypothalamus is treated safely andefficiently by GKS. Type II HH can be resected by either endoscopic or transcallosalapproaches or treated by GKS depending on the parent’s choice and severity of epilepsy.In small type III HH, GKS is the safer procedure because of the very close relationship tothe fornix and mammillary bodies. Types V (rarely epileptic) and IV are frequently operableby disconnection. Very large type VI (or mixed type) with a large component above the floorof the third ventricle must be disconnected, and then the upper remnant is best treated byGKS using a staged technique. Overall, when the lesion is sufficiently small, GKS offers arate of seizure control comparable to microsurgery but with much lower risk. The disad-vantage of radiosurgery is its delayed action. Longer follow-up is mandatory for a reliableevaluation of the role of GKS.Semin Pediatr Neurol 14:73-79 © 2007 Elsevier Inc. All rights reserved.

KEYWORDS hypothalamic hamartoma, gamma knife surgery, resective microsurgery, discon-nective surgery, epilepsy, poikilothermia

Hypothalamic hamartomas (HHs) are rare congenital het-erotopic lesions that are intrinsically epileptogenic

when closely connected to the mammillary bodies.1,2 Patients

classically present with gelastic seizures during the first yearsof life.3 In the more severe cases, affected patients develop anepileptic encephalopathy during the following years,4 char-acterized by drug resistance, various types of seizures withgeneralisation (including drop attacks),3 cognitive decline,5-8

and severe psychiatric comorbidity.9 Usually, seizure semiol-ogy suggests the involvement of temporal or frontal loberegion and suggests a phenomenon of secondary epilepto-genesis.10-12

In 1969, Paillas and coworkers1 first showed that the epi-lepsy associated with HH, including the signs of involvementof temporal or frontal lobe regions, can be alleviated withsurgical resection of the HH lesion itself. Direct proof of therole of the HH in generating seizure activity was provided by

*Department of Functional Neurosurgery, INSERM 751, Timone Hospital,Marseilles, France.

‡Department of Neurophysiology, INSERM 751, Timone Hospital, Mar-seilles, France.

†Department of Pediatric Neurosurgery, Timone Hospital, Marseille,France.

§Department of Endocrinology, Timone Hospital, Marseille, France.&Department of Psychiatry, Timone Hospital, Marseille, France.Address reprint requests to Jean Régis, MD, Service de Neurochirurgie Fonc-

tionnelle et Stéréotaxique, C.H.U. La Timone, 264 rue Saint Pierre,13,385 Marseille, Cedex 05. E-mail: jregis@ap-hm.fr

731071-9091/07/$-see front matter © 2007 Elsevier Inc. All rights reserved.doi:10.1016/j.spen.2007.03.005

Munari et al,12 relying on data obtained from stereo encepha-lographic recordings from implanted depth electrodes.13,14

This evidence was reinforced by ictal single photon emissioncomputed tomography (SPECT) studies15-17 and other stud-ies using depth electrodes.5,18

A large spectrum of surgical techniques and approachesattempting to directly treat the HH has been proposed. Dur-ing recent years, we have tried to evaluate precisely the ben-efit of the various modalities of treatment to better determinethe most appropriate one for a given child and to be able toprovide the parents balanced and validated information.19

We have been interested in the specific role of gamma kniferadiosurgery in this special group of children.

RadiosurgeryBy definition, radiosurgery is a neurosurgical procedure inwhich converging narrow ionizing beams are stereotacticallyfocused to induce a desired biological effect in a predeter-mined target, with minimal radiation to the surrounding tis-sues and without opening the skull.

Enormous experience has been gained worldwide usinggamma knife surgery (GKS) in various indications, showingthat the side effects of radiosurgery are rare, generally tran-sient, and quite easily predictable.20 GKS must be consideredas an option for small, deeply seated lesions in which thereare risks of complications or functional worsening with re-sective surgery. For these indications, GKS compares favor-ably with microsurgical removal in terms of safety, efficacy,and also cost-effectiveness.

The use of GKS in various pathological conditions associ-ated with severe epilepsy has revealed a likely specific anti-epileptic effect.21 For example, in a homogeneous group ofpatients with mesial temporal lobe epilepsies, we have shownthe short- and intermediate-term safety and efficacy GK sur-gery.22,23

In HHs, because of the intrinsic epileptogenicity of thisrelatively well-circumscribed lesion1,13,24 radiologic targetingis dramatically simplified. Arita and coworkers25 in 1999reported the first case of successful treatment by gamma kniferadiosurgery, with no side effects. We published in 2000 thefirst series showing the very good safety profile and efficacy ofradiosurgery in patients with HH and refractory epilepsy.26

The excellent results observed in this retrospective series of 8patients (all improved, 50% cured of seizures, and no adverseeffects except 1 case with transient poikilothermia) led us toorganize a prospective multicenter trial. Several case reportsin the medical literature are in good accordance with theresults of our retrospective study.27,28 Because of the relativerarity of this disease, most earlier clinical series of HH treat-ment that have been reported have been extremely small(typically from 1 to 5 cases), with the exception of the retro-spective multicenter series of Palmini and coworkers,29

which aggregated 13 cases. Our current series of 60 prospec-tively evaluated patients is unique by virtue of the number ofpatients and the strict methodology of evaluation.19

Efficacy of Radiosurgeryfor HH and Refractory EpilepsyWe have published a preliminary report of this study.19 Thirty-one patients were treated with GKS at least 3 years before thisreport, with satisfactory information regarding posttreatmentfollow-up available for 27 patients. The preoperative evaluationincluded detailed assessment of cognitive deficits and behavioraldisturbances and investigated the relationship between seizureseverity and anatomic HH type and cognitive abilities.6,9 Thegoal of the preoperative workup was to adequately select thecandidates for inclusion and to evaluate baseline neurologic andendocrinologic function.

All radiosurgical procedures were performed by using theLeksell 201-source Cobalt 60 Gamma Knife (Elekta Instru-ment, Stockholm, Sweden).2,30,31 We used a multi-isocentriccomplex dose protocol with high conformity and selectivity.We used low peripheral doses to take into account the closerelationship with radiosensitive optic pathways and hypo-thalamus (median, 17 Gy at the 50% isodose margin; range,13 to 26 Gy). The lesions treated in this series were generallysmall (median diameter, 9.5 mm; range, 5-26 mm). We paidspecial attention to the dose delivered to the mammillarybody and to the fornix, and we always tailor the dose plan foreach patient, based on the use of a single run of shots with the4-mm collimator.

Patient outcome was evaluated with respect to seizures,cognition, behavior, and endocrine status 6, 12, 18, 24, and36 months after radiosurgery and then once yearly. Amongthe study group of 27 patients, 10 are seizure free (37%), and6 very much improved (22.2%) with a significant reductionin seizure frequency (usually only rare gelastic seizures), as-sociated with dramatic behavioral and cognitive improve-ment. Overall, very good results were obtained in 16 (60%)of the patients. Five patients (18.5%) with small hamartomasare marginally improved and are under consideration forrepeat radiosurgery. Two patients have thus far reported nosignificant improvement. According to our protocol, the pa-tient and the family are offered a second radiosurgery ininstances of partial benefit when the lesion is anatomicallysmall and well defined. The radiosurgical treatment was per-formed twice in 9 patients. A microsurgical resection wasperformed in 4 patients (14.8%) with large HH and poorefficacy of radiosurgery. After this microsurgical approach, 2are seizure free, and 2 are improved but still experiencingnumerous seizures.

Effect of GKS on Behaviorand Cognitive FunctionNine patients in this cases series (33.3%) reported dramaticimprovement. All patients with paroxysms of aggressivenessimproved substantially. Increased alertness, elevated mood,and greater speech production were observed in some pa-tients previously characterized by excessive behavioral inhi-bition. The parents, primarily in the younger patients, fre-quently reported a positive effect on sleep patterns. Finally,

74 J. Régis et al

dramatic developmental acceleration was observed in 3young patients.

Evolution of the Therapeutic ResponseThe pattern of the time course of postoperative effectivenesson seizures was similar in the majority of patients, althoughthe various effects of the treatment could be observed withdifferent absolute time points in different subjects. The typ-ical pattern comprises 5 periods extending over a minimumof 2 years. At first, patients experience generally a quite im-mediate slight global improvement, with reduced seizure fre-quency and severity. Usually, an improvement in behavior isalso observed during this phase. During the followingmonths (typically postoperative months 2-6), the frequencyof seizures is back to the preoperative baseline. In this secondperiod, despite a constant number of seizures, other param-eters start to improve including background encephalogramand sleep normalize, behavior improves, the capacity to gainnew skills at school is enhanced, and the overall quality of lifeimproves.

The third period often sets in suddenly, marked by anincreased number of seizures, sometimes with a large num-ber over only several days. The duration of this period is brief,days or weeks, and rarely more than 1 month. The fourthperiod is marked by the progressive disappearance of sei-zures. The fifth period is the consolidation period. Interindi-vidual variability exists mainly in the delay of onset and du-ration of these periods and in the amplitude of the clinicalphenomenon described previously. However, knowledge ofthis pattern obtained from typical patients is very helpful forestablishing the expectations for more atypical patients inwhom these phases are present but are generally more diffi-cult to discern.

ComplicationsThe major advantage of radiosurgery is its exceptional safety.Thus far, no permanent neurologic deficits have ever beenreported in an HH patient with this treatment modality.From the radiologic point of view, the majority of our pa-tients did not have any obvious change in MRI (28/31 pa-tients, 90.3%) and did not exhibit even transient side effectsof GKS treatment.19 MRI scans performed at 6, 12, 18, 24,and 36 months have usually shown the absence of anychanges even on the fluid attenuated inversion recovery se-quences. In 2 patients, a slight decrease in the volume of theHH lesion was observed (without increased signal on T2-weighted images). In 2 other patients, high T2-weighted sig-nal was transiently visible in the HH lesion and in the imme-diately adjacent hypothalamus, without mass effect orclinical signs or symptoms.

Only 4 patients (14.8%) have experienced a transientworsening of the epilepsy and 3 (11%) a transient poikilo-thermia. None of the patients in this series has presented apermanent complication. Specifically, we have not observedappetite stimulation or weight gain, diabetes insipidus, hy-pothyroidism, or other endocrinologic complications. No pa-

tient has complained of a worsening of cognitive abilities,and, in particular, we have not observed any complaintsabout short-term memory function. No new visual acuity orvisual field deficits have been observed.

Possible long-term risks are an especially important issuein children. To date, rare cases of patients treated with radio-surgery for other conditions have been reported in which anew malignant tumor, close to the initial target of radiosur-gery, was discovered several years after treatment. A cause-and-effect relationship between radiosurgery and the newtumor appears unlikely in several of these cases and a matterof debate in others. However, even if all these cases wereaccepted as caused by radiosurgery, the risk would still ap-pear to be very low (between 1 case in 10,000 to 100,000treated patients) according to Lunsford et al.32 This risk is farlower in comparison to the risk of death after microsurgicalresection (1%-2%) and appears to be acceptable in light ofthe often catastrophic natural history of untreated HH. How-ever, parents should be systematically informed of the poten-tial risk of secondary tumors after GKS.

Comparison to OtherNeurosurgical StrategiesRadiosurgery is undoubtedly the least invasive strategy fordestructive or ablative therapy of HH associated with intrac-table epilepsy.2,19,25-28,33-35 The vascular risk of microsurgicalapproaches and stereotactic thermocoagulation16,36,37 havenever been reported with radiosurgery. Concerns regardingthe impairment of short-term memory in children who arecognitively intact38,39 is promoting the exploration of lessinvasive alternatives such as radiosurgery. Thus, it bears re-peating that none of the patients in our series have com-plained of posttreatment memory deficits.

Radiosurgery as an image-guided procedure improves thedelineation of the boundary between the abnormal tissue ofthe HH and the adjacent normal hypothalamus as comparedwith resective surgery using the endoscopic or transcallosalapproach.38-40 GKS allows more precise definition of the do-simetry contour map in comparison to interstitial brachy-therapy,19 and the rate of seizure freedom after treatmentappears superior to brachytherapy.41 Additionally, there is a1% to 2% risk of hemorrhage associated with insertion ofstereotactic probes into the brain, a necessary feature of in-terstitial radiotherapy.

The disconnection strategy through a pterional approach,proposed by Delalande and Fohlen,40 allows treatment to thepart of the HH lesion inferior to the floor of the third ventriclewith less risk compared with bulk resection, requiring dis-section in the cisterns, through the same approach. The tran-scallosal interforniceal resection is a very attractive approachfor HH lesions with an intraventricular component, includ-ing types I, II, and III. Surgical resection series, using thetranscallosal approach, from both the Barrow NeurologicalInstitute in Phoenix42 and Royal Children’s Hospital in Mel-bourne43-45 report very good results with 54% and 52%, re-spectively, of the patients seizure free postoperatively. The

Gamma knife surgery 75

major problem of this approach is the frequent injury to thefornices and the associated risk of short-term memory im-pairment.38,39

Consequently, a transventricular endoscopic approach forHH surgery has been proposed.39,40,46,47 Rekate and cowork-ers39 have recently reported preliminary results on a series of44 patients with HH and refractory epilepsy. Of the 14 pa-tients with 100% resection of the HH lesion, 13 were seizurefree. The majority of these lesions were less than 1 cm3

(72%). It is of interest to note that a complete removal wasachieved in only 14 patients (31.8%). The same authors havepreviously shown that seizure outcomes relate directly to thesurgeon’s ability to remove the entire hamartoma.42 Clinicalcomplications occurred in 11 patients (25%) but were de-scribed as permanent in only 3 (6.8%).39 Thalamic infarctionwas revealed by postoperative diffusion-weighted MRI in 11cases (25%). Of these, 8 were asymptomatic, and 3 experi-enced hemiparesis with complete resolution in 2 and persis-tence in 1. Thus, the total complication rate was 43% (19patients). Among the patients with clinical complications, 6had short-term memory impairment, including 3 still pre-senting with this deficit at the 3-month follow-up timepoint.39 The disconnection strategy and the transcallosal in-terforniceal approaches are, with GKS, the 3 most interestingin terms of safety and efficacy.19

Topological HHClassification andTreatment StrategyOur experience with this group of patients has led us toconsider that the surgical strategy must be based on both thetopological type of the lesion2,19 and the clinical severity ofthe clinical manifestations. The severity of the seizure disor-der determines the timing of surgical treatment and some-times may also influence the choice of the approach. Whenthe frequency of seizures is high, the epileptic encephalopa-thy associated with HH must be prevented by early surgery.Optimally, intervention must occur before the beginning ofthe mental decline and deterioration of seizure control. Insuch instances, the delayed efficacy of radiosurgery can be amajor limitation.48 On the other hand, the excellent safetyprofile of radiosurgery may allow treatment to be plannedproactively and to propose a very early curative surgical ap-proach before clinical deterioration.

The topological classification of the HH lesion based on agood high-resolution MRI study is a key feature in the deci-sion-making process.2 Previous classification systems havebeen proposed based on anatomic49,50,51 or surgical40 consid-eration. These classification systems fail to describe the largediversity of HH lesions and the treatment considerations as-sociated with this diversity. As emphasized by Palmini andcoworkers,29 the exact location of the lesion in relation to theinterpeduncular fossa and the walls of the third ventriclecorrelates with the extent of excision, postoperative seizurecontrol, and complication rate.

On this basis, we have proposed an anatomical classifica-

tion system for HH lesions (Fig 1).2,19 In our experience, thisclassification correlates with the clinical symptoms and sei-zure severity and is especially critical for deciding on surgicalstrategy. Type I HH (small HHs located inside the hypothal-amus extending more or less into the third ventricle) is cer-tainly the best candidate for GKS. In this population, the riskof microsurgical removal is likely to be potentially high. Intype II HH (when the HH lesion is small and is mainly withinthe third ventricle), radiosurgery is certainly the safer alter-native. Even though the endoscopic and transcallosal inter-forniceal approaches have been proposed, the risks of short-term memory impairment, endocrinological disturbance(including hyperphagia with resulting obesity, hypothyroid-ism, or diabetes insipidus), and thalamic or thalamocapsularinfarction have been reported by these enthusiastic and skil-ful neurosurgeons. However, in cases of very severe repeatedepisodes of status epilepticus, we propose as a salvage surgeryeither the transcallosal interforniceal approach or the trans-ventricular endoscopic approach (depending on the width ofthird ventricle). In emergency situations, if the HH lesion issmall and the third ventricle is large, the endoscopic ap-proach is chosen.

In type III (the HH lesion is located essentially in the floorof the third ventricle), the extremely close relationship be-tween the mammillary body, the fornix, and the HH clearlyleads us to prefer GKS. We speculate that sessile HHs (ie,those associated with epilepsy) always have an anatomic con-nection in the hypothalamus close to, or on, the mammillary

Figure 1 Topological classification of HHs.19 Classification of HHsinto these 6 categories based on MRI findings resulted in a clearcorrelation between symptoms and the subsequent clinical courseand is pertinent for clinical management. (Reprinted with permis-sion.19)

76 J. Régis et al

body. Thus, even a lesion classified as type II, which meansthat it appears on the MRI primarily within the third ventri-cle, is likely to have a base of connection in the hypothala-mus.

In type IV (a sessile lesion within the interpeduncular cis-tern), an open surgical disconnection can be discussed, suchas a pterional approach, with or without orbitozygomaticosteotomy. However, if the lesion is small, GKS can be rec-ommended because of its safety profile and its capability toreach at the same time the small part of the lesion within thehypothalamus itself, frequently visible on high-resolutionMRI. In Delalande and Fohlen’s experience, only 2 patientsamong 14 are seizure free after a single disconnectionthrough a pterional approach.40 Consequently, we do usethis approach in the case of a lesion too large for GKS as a firststep of a staged approach. In most circumstances, the patientis improved but not seizure free after the first surgical step,and GKS is performed 3 months after the first procedure as asecond step of the treatment.

Type V HHs (pedunculated HH lesions connected to theinferior aspect of the hypothalamus by a clear stalk) are rarelyepileptic. These lesions typically result in central precociouspuberty without epilepsy or significant neurobehavioralproblems and are treated medically with gonadotropin-re-leasing hormone agonists. However, if surgery is required,they can easily be treated by radiosurgery or disconnectionthrough a pterional approach. In case of severe epilepsy, thesecond therapeutic modality will certainly allow for a fastertherapeutic response. However, a distant extension of the HHinto the hypothalamus, close to the mammillary bodies, mustbe cautiously searched for on high-resolution MRI. Its dis-covery leads us to prefer GKS, which enables treatment toboth parts of the lesion, especially in cases in which thecisternal component is small.

Type VI (giant HH lesions) require a combination of ther-apeutic modalities (a staged approach) in nearly all instances.In our experience, efforts to functionally disconnect the HHlesion with radiosurgery have been consistently disappoint-ing. In our opinion, this strategy may cause loss of precioustime for the child to be treated effectively. Consequently, wedo not advocate for such a strategy. When microsurgicalresection has left a small remnant in the third ventricle andstill active epilepsy, reoperation can be envisaged by GKS.

For retreatment after a first GKS procedure, the key pointis to take into account the previously observed phenomenonof delayed therapeutic effectiveness and the probability ofseizure cessation at different points in time after GKS. Be-cause of the existence of interindividual variability (ie, thepossibility of late responders), we recommend waiting for aminimum of 36 months after the first radiosurgery treatmentbefore proposing a second treatment. The excellent safety ofgamma knife radiosurgery and the history of successful re-sults with re-treatment is a strong argument to propose asecond radiosurgical treatment step in patients with persist-ing seizures, particularly those with reduced seizure fre-quency after the first round of GKS and a small-size HH.There are few doubts that gamma knife radiosurgery is theleast invasive approach we can propose for these children,

with comparable efficacy to resective procedures. The dog-matic approach of supporting only 1 modality of HH therapyover all others is to be discouraged because it fails to take intoaccount the anatomic and clinical heterogeneity of this pa-tient population.

A further question concerns the existence of preoperativepredictors of success. In our retrospective study, only treat-ment at an earlier age, when epilepsy was not yet too severe,was associated with a high chance of seizure control afterradiosurgery. In this group, those with successful seizurecontrol had significantly fewer gelastic seizures before radio-surgery (median, 6 gelastic seizures/month) compared withthe group with continuing seizures (median 45 gelastic sei-zures/month).26 Preoperatively, the ability to deliver optimaldoses at the lesion margin for small type I, II, or III HH lesionswas a positive predictor of therapeutic response. Accordingto the Melbourne team, a less severe form of epilepsy, as wellas relatively normal cognitive function, were positive predic-tors of good outcome in terms of the likelihood of completeseizure freedom.43

The role of secondary epileptogenesis or, alternatively, ofwidespread cortical dysgenesis needs to be better evaluatedand understood to identify the best indications and the besttiming for the different surgical approaches.

ConclusionIn an individual patient presenting with HH and refractoryepilepsy, the choice of the therapeutic strategy must be basedboth on the precise topological classification and on the se-verity of the clinical manifestations. The age at the time ofradiosurgery (or resective surgery) has a major influence onboth the probability of seizure cessation and reversal of theepileptic encephalopathy. Consequently, in severe cases,early surgical intervention is probably to be recommended inan attempt to minimize or prevent the cognitive and behav-ioral problems commonly seen with this epileptic syndrome.Radiosurgery can alleviate both the seizures and the behav-ioral and cognitive abnormalities associated with HH. Un-doubtedly, radiosurgery is the least invasive curative tech-nique we can propose for such lesions. Complications arevery rare, especially when compared with microsurgery. GKSis not indicated in giant HH lesions (stage VI) or in patientswith extremely severe psychiatric disorder. Because of theseverity of the clinical prognosis in the majority of thesepatients and the potential for complications with microsur-gical resection, GKS can presently be envisaged as the firsttherapy of choice for small-sized and midsized HH lesionsassociated with epilepsy. It may improve the future of theseyoung patients dramatically. In large HH lesions, a 2-stagetreatment strategy can be proposed, including a microsurgi-cal disconnection of the lower part, followed by radiosurgeryof the intraventricular remnant.

AcknowledgmentsWe thank the Assistance Publique Hôpitaux de Marseillewho have actively supported this work.

Gamma knife surgery 77

References1. Paillas JE, Roger G, Toga M, et al: Hamarthome de l’hypothalamus. Rev

Neurol 120:177-194, 19692. Régis J, Hayashi M, Perez Eupierre L, et al: Gamma knife surgery for

epilepsy related to hypothalamic hamartomas. Acta Neurochir 91:33-50, 2004

3. Tassinari CA, Riguzzi P, Rizzi R, et al: Gelastic seizures, in Tuxhorn I,Holthausen H, Boenigk H (eds): Current Problems in Epilepsy (vol 1).London: John Libbey, 1997, pp 429-446

4. Kerrigan JF, Ng YT, Chung S, et al: The hypothalamic hamartoma: Amodel of subcortical epileptogenesis and encephalopathy. Semin Pedi-atr Neurol 12:119-131, 2005

5. Berkovic SF, Andermann F, Melanson D, et al: Hypothalamic hamar-toma and associated ictal laughter. Evolution of a characteristic epilep-tic syndrome and diagnostic value of magnetic resonance imaging. AnnNeurol. 23:429-439, 1998

6. Frattali CM, Liow K, Craig GH, et al: Cognitive deficits in children withgelastic seizures and hypothalamic hamartoma. Neurology 57:43-46, 2001

7. Nguyen D, Singh S, Zaatreh M, et al: Hypothalamic hamartomas: Sevencases and review of the literature. Epilepsy Behav 4:246-258, 2003

8. Quiske A, Frings L, Wagner K, et al: Cognitive functions in juvenile andadult patients with gelastic epilepsy due to hypothalamic hamartoma.Epilepsia 47:153-158, 2006

9. Weissenberger AA, Dell ML, Liow K, et al: Aggression and psychiatriccomorbidity in children with hypothalamic hamartomas and their unaf-fected siblings. J Am Acad Child Adolesc Psychiatry 40:696-703, 2001

10. Freeman JL, Coleman LT, Wellard RM, et al: MR imaging and spectro-scopic study of epileptogenic hypothalamic hamartomas: analysis of 72cases. AJNR Am J Neuroradiol 25:450-462, 2004

11. Mullatti N, Selway R, Nashef L, et al: The clinical spectrum of epilepsyin children and adults with hypothalamic hamartoma. Epilepsia 44:1310-1319, 2003

12. Munari C, Kahane P, Francione S, et al: Role of the hypothalamichamartoma in the genesis of gelastic fits (a video-stereo-EEG study).Electroencephalogr Clin Neurophysiol 95:154-160, 1995

13. Kahane P, Tassi L, Hoffmann D, et al: Crises dacrystiques et hamartomehypothalamique: Àpropos d’une observation vidéo-stéréo-EEG. Epi-lepsies 6:259-279, 1994

14. Tassi L, Quarato P, Francione S: Chronological relationships betweenictal laughing and smiling and intralesional discharges: Stereo-EEGstudy of a patient with gelastic epilepsy and hypothalamic hamartoma.Epilepsia 36:239-240, 1995

15. Donley D, Kuzniecky R, Mountz J, et al: Ictal SPECT findings in hypo-thalamic hamartoma and epilepsy. Epilepsia 35:146, 1994 (suppl 8,abstr)

16. Kuzniecky R, Guthrie B, Mountz J, et al: Intrinsic epileptogenesis ofhypothalamic hamartomas in gelastic epilepsy. Ann Neurol 42:60-67,1997

17. Kuzniecky R, Guthrie B, Mountz J, et al: Hypothalamic hamartomasand gelastc seizures: Evidence for subcortical seizure generation by ictalSPECT and cerebral stimulation. Epilepsia 36:s266, 1995 (suppl 3)

18. Arroyo S, Santamaria J, Sanmarti F, et al: Ictal laughter associated withparoxysmal hypothalamopituitary dysfunction. Epilepsia 38:114-117,1997

19. Regis J, Scavarda D, Tamura M, et al: Epilepsy related to hypothalamichamartomas: Surgical management with special reference to gammaknife surgery. Childs Nerv Syst 22:881-895, 2006

20. Flickinger J, Kondziolka D, Lunsford L: Clinical applications of stereo-tactic radiosurgery. Cancer Treat Res 93:283-297, 1998

21. Regis J, Bartolomei F, Hayashi M, et al: The role of gamma knife surgeryin the treatment of severe epilepsies. Epileptic Disord 2:113-122, 2000

22. Regis J, Rey M, Bartolomei F, et al: Gamma knife surgery in mesialtemporal lobe epilepsy: A prospective multicenter study. Epilepsia 45:504-515, 2004

23. Régis J, Bartolomei F, Rey M, et al: Gamma knife surgery for mesialtemporal lobe epilepsy. Epilepsia 40:1551-1556, 1999

24. Plouin P, Ponsot G, Dulac O, et al: Hamartomes hypothalamiques etcrises de rire. Rev Electroencephalogr Neurophysiol Clin 13:312-316,1983

25. Arita K, Kurisu K, Iida K, et al: Subsidence of seizure induced bystereotactic radiation in a patient with hypothalamic hamartoma. Casereport. J Neurosurg 89:645-648, 1998

26. Regis J, Bartolomei F, de Toffol B, et al: Gamma knife surgery forepilepsy related to hypothalamic hamartomas. Neurosurgery 47:1343-1351, 2000

27. Barajas MA, Ramirez-Guzman MG, Rodriguez-Vazquez C, et al:Gamma knife surgery for hypothalamic hamartomas accompanied bymedically intractable epilepsy and precocious puberty: experience inMexico. J Neurosurg 102:53-55, 2005

28. Unger F, Schrottner O, Haselsberger K, et al: Gamma knife radiosur-gery for hypothalamic hamartomas in patients with medically intracta-ble epilepsy and precocious puberty. Report of two cases. J Neurosurg92:726-731, 2000

29. Palmini A, Chandler C, Andermann F, et al: Resection of the lesion inpatients with hypothalamic hamartomas and catastrophic epilepsy.Neurology 58:1338-1347, 2002

30. Leksell L: The stereotaxic method and radiosurgery of the brain. ActaChir Scand 102:316-319, 1951

31. Leksell L: Stereotaxis and radiosurgery. An operative system, inThomas CC (ed): Stereotaxis and Radiosurgery. An Operative System.Springfield, IL, 1971

32. Lunsford LD, Niranjan A, Flickinger JC, et al: Radiosurgery of vestib-ular schwannomas: Summary of experience in 829 cases. J Neurosurg102:195-199, 2005 (suppl 1)

33. Dunoyer C, Ragheb J, Resnick T, et al: The use of stereotactic radiosur-gery to treat intractable childhood partial epilepsy. Epilepsia 43:292-300, 2002

34. Mathieu D, Kondziolka D, Niranjan A, et al: Gamma knife radiosurgeryfor refractory epilepsy caused by hypothalamic hamartomas. StereotactFunct Neurosurg 84:82-87, 2006

35. Ogino W, Soma O. [A case of hypothalamic hamartoma with refractoryseizures: Successful treatment with gamma knife surgery]. No To Hat-tatsu 35:65-69, 2003

36. Fujimoto Y, Kato A, Saitoh Y, et al: Stereotactic radiofrequency ablationfor sessile hypothalamic hamartoma with an image fusion technique.Acta Neurochir (Wien) 145:697-700, 2003

37. Parrent AG: Stereotactic radiofrequency ablation for the treatment ofgelastic seizures associated with hypothalamic hamartoma. Case re-port. J Neurosurg 91:881-884, 1999

38. Maixner W: Hypothalamic hamartomas—Clinical, neuropathologicaland surgical aspects. Childs Nerv Syst 22:867-873, 2006

39. Rekate HL, Feiz-Erfan I, Ng YT, et al: Endoscopic surgery for hypotha-lamic hamartomas causing medically refractory gelastic epilepsy.Childs Nerv Syst 22:874-880, 2006

40. Delalande O, Fohlen M: Disconnecting surgical treatment of hypothalamichamartoma in children and adults with refractory epilepsy and proposal ofa new classification. Neurol Med Chir (Tokyo) 43:61-68, 2003

41. Schulze-Bonhage A, Homberg V, Trippel M, et al: Interstitial radiosur-gery in the treatment of gelastic epilepsy due to hypothalamic hamar-tomas. Neurology 62:644-647, 2004

42. Ng Y-T, Rekate HL, Prenger EC, et al: Transcallosal resection of hypotha-lamic hamartoma for intractable epilepsy. Epilepsia 47:1192-1202, 2006

43. Rosenfeld JV, Freeman JL, Harvey AS: Operative technique: The ante-rior transcallosal transseptal interforniceal approach to the third ven-tricle and resection of hypothalamic hamartomas. J Clin Neurosci 11:738-744, 2004

44. Rosenfeld JV, Harvey AS, Wrennall J, et al: Transcallosal resection ofhypothalamic hamartomas, with control of seizures, in children withgelastic epilepsy. Neurosurgery 48:108-118, 2001

45. Harvey AS, Freeman JL, Berkovic SF, et al: Transcallosal resection ofhypothalamic hamartomas in patients with intractable epilepsy. Epi-leptic Disord 5:257-265, 2003

46. Akai T, Okamoto K, Iizuka H, et al: Treatments of hamartoma withneuroendoscopic surgery and stereotactic radiosurgery: A case report.Minim Invasive Neurosurg 45:235-239, 2002

47. Choi JU, Yang KH, Kim TG, et al: Endoscopic disconnection for hypo-thalamic hamartoma with intractable seizure. Report of four cases.J Neurosurg 100:506-511, 2004

78 J. Régis et al

48. Ng YT, Rekate HL: Emergency transcallosal resection of hypotha-lamic hamartoma for “status gelasticus.” Epilepsia 46:592-594,2005

49. Arita K, Ikawa F, Kurisu K, et al: The relationship between magneticresonance imaging findings and clinical manifestations of hypotha-lamic hamartoma. J Neurosurg 91:212-220, 1999

50. Debeneix C, Bourgeois M, Trivin C, et al: Hypothalamic hamartoma:Comparison of clinical presentation and magnetic resonance images.Horm Res 56:12-18, 2001

51. Valdueza JM, Cristante L, Dammann O, et al: Hypothalamic hamarto-mas: With special reference to gelastic epilepsy and surgery. Neurosur-gery 34:949-958, 1994

Gamma knife surgery 79