Acta Neurol Belg DOI 10.1007/s13760-014-0374-z
ORIGINAL ARTICLE
Acute intralesional recording in hypothalamic hamartoma: description of 4 cases Nicola Specchio • Michele Rizzi • Marina Trivisano • Lucia Fusco • Erica Rebessi • Simona Cappelletti • Luca De Palma • Flavio Villani • Alessandra Savioli • Alessandro De Benedictis • Carlo Efisio Marras • Federico Vigevano • Olivier Delalande
Received: 21 July 2014 / Accepted: 30 September 2014 Ó Belgian Neurological Society 2014
Abstract Hypothalamic hamartomas (HHs) are intrinsically epileptogenic lesions associated to medically intractable focal epilepsy mainly characterized by gelastic and focal seizures. Intralesional recording with deep electrodes has documented the presence of ictal discharge arising from inside the lesion. Nevertheless interictal and ictal scalp EEG is poorly informative and non-localizing in a great deal of cases. HH disconnection leads to seizure remission in most cases. To describe the intralesional EEG recordings and to compare them with concomitant scalp EEG and with previous cases reported in literature. We reviewed the medical records of 17 children affected by drug-resistant focal epilepsy associated to HH. We recorded intralesional electrical activity during stereo-endoscopic disconnection in three cases and during deep brain stimulation implantation in one. We also correlated it with
Electronic supplementary material The online version of this article (doi:10.1007/s13760-014-0374-z) contains supplementary material, which is available to authorized users. N. Specchio (&) L. Fusco L. De Palma F. Vigevano Division of Neurology, Department of Neuroscience, Bambino Gesu` Children’s Hospital IRCCS, P.za S. Onofrio, 4 00165 Rome, Italy e-mail: [email protected] M. Rizzi Division of Neurosurgery, IRCCS Neurological Institute ‘‘C.Besta’’, Milan, Italy M. Trivisano Clinic of Nervous System Diseases, University of Foggia, Foggia, Italy E. Rebessi A. De Benedictis C. E. Marras O. Delalande Division of Neurosurgery, Department of Neuroscience, Bambino Gesu` Children’s Hospital IRCCS, Rome, Italy
the simultaneous scalp-EEG recording. Acute intralesional recordings in our cases confirmed the presence of epileptiform abnormalities intermingled with low-voltage activity, mostly on the same side of the HH attachment. Paroxysmal activity recorded inside the HH was always evident. Mapping of HH epileptogenic activity could be useful to confirm the usefulness of disconnection procedure. This should consider on-site recording from the HH and if abnormalities are detected safely proceed to disconnection of the HH. Keywords Hypothalamic hamartoma Gelastic seizures Depth electrodes Invasive monitoring Epilepsy surgery
Introduction Hypothalamic hamartomas (HHs) are well-recognized congenital malformations of the brain, defined as benign, tumor-like nodules of disorganized but mature cells mostly S. Cappelletti Psychology Unit, Department of Neuroscience, Bambino Gesu` Children’s Hospital IRCCS, Rome, Italy F. Villani Division of Clinical Epileptology and Experimental Neurophysiology, Fondazione Istituto Neurologico ‘‘Carlo Besta’’, Milan, Italy A. Savioli Division of Anaesthesia and Emergency Room, Bambino Gesu` Children’s Hospital IRCCS, Rome, Italy O. Delalande Service de Neurochirurgie Pe´diatrique, Fondation Ophtalmologique Adolphe de Rothschild, Paris, France
123
Acta Neurol Belg
composed of a combination of neuronal or ganglion cells, glial cells and blood vessels [1]. HH can be responsible for neurological and endocrine disturbances. HH represent a cause of medically intractable focal epilepsy. Antiepileptic drugs (AEDs) are poorly effective in patients with HH [2, 3] therefore early surgical treatment should be considered in every case [4, 5]. Two different surgical approaches, resection or disconnection, can lead to remission of the attacks, even after AED withdrawal [6, 7]. There is also evidence that radiofrequency and Gamma knife radiosurgery may lead to a complete seizure control with limited morbidity, when the entire lesion can be targeted [8–11]. HH produces stereotyped epileptic seizures characterized by paroxysmal laughing or crying (gelastic or dacrystic seizures), other than focal seizures with or without motor manifestations [2]. Seizures onset is usually in the first months of life and frequency is variable from daily to weekly recurrence. Interictal and ictal scalp EEG is usually poorly informative in most cases: focal or diffuse spike and wave complexes have been recorded in some cases in some cases, focal theta and delta activity could also be associated with epileptiform abnormalities [12]. Ictal EEG usually shows diffuse flattening or focal theta rhythmic activity. Interictal abnormalities and ictal changes usually predominate over the side ipsilateral to the hamartoma [13]. Intralesional recording with an electrode implanted in the hamartoma showed the presence of an ictal discharge arising inside the lesion associated to the clinical pattern [13, 14]. Within a collaborative epilepsy surgery program we studied 17 patients with drug-resistant epilepsy associated to HH. Acute intralesional recording of the interictal activity arising inside the HH was performed during deep brain stimulation (DBS) electrode implantation, and endoscopic disconnections. The aim of this report is to describe the results of intralesional recordings, to compare them with concurrent scalp-EEG recordings and with previous cases reported in literature.
Methods We have reviewed the medical records of 17 children affected by drug-resistant focal epilepsy associated to HH, referred between January 1990 and December 2011 to the Neurology Division of the Bambino Gesu` Children’s Hospital in Rome and ‘‘Carlo Besta’’ Neurological Institute in Milan: both places receive tertiary patients for specialist consultations. All clinical notes were reviewed to evaluate the age at onset, the seizure type at onset and during the follow-up,
123
the treatment and the epilepsy outcome. Seizures were defined according to the Glossary of Descriptive Terminology for Ictal Semiology [15]. HHs were defined according to the classification proposed by Delalande [4]. Mean patient age at seizure onset was 4,67 ±3.9 years (range 1 day–13 years). At onset 14 patients presented with gelastic seizures, 1 patient with facial grimace, 1 patient with hemifacial spasms associated to ocular blinking, and 1 with ictal syncope. During the follow-up other type of seizures appeared: generalized tonic–clonic (3 pts), tonic (1 pt), atonic (1 pt), clonic (1 pt), and dacrystic (1 pt). In 10 out of 17 cases a surgical approach was performed: 4 of them underwent a one stage stereo-endoscopic disconnection, 3 cases a double stage stereo-endoscopic disconnection, and 1 case required a multistage stage stereoendoscopic disconnection, 1 patient performed a DBS protocol, and 1 a surgical resection. Moreover, 2 patients were treated with radiotherapy, and radiosurgery. Following Delalande classification [4], HHs were mostly of type 2 (8 out of 17), followed by type 3 (6 out of 17) and type 4 (3 out of 17). No cases were classified as type 1. See Supplementary Table 1. In four patients an intraoperative endoscopic intralesional recording was performed. In these cases we also collected data regarding pre- and post-operative neurological, neuropsychological and endocrinological evaluations, EEG findings including interictal and ictal long-term monitoring; moreover, a detailed treatment history was reported. Clinical and neurophysiological data of patient 1 have already been reported by Marras et al. [16]. In 3 cases intraoperative stereo EEG was carried out using foramen ovale electrodes (Dixi Medical ACS-798S, 5 contacts), and DBS leads in 1 case (Medtronic 3389, 4 contacts: 1.27 mm diameter, 1.5 mm long, spaced by 0.5 mm). Foramen ovale electrodes were introduced in the core of the hamartomas and activity was recorded for at least 10 min. For the foramen ovale electrodes only the first 4 contacts were inserted and activity was recorded in referential montage, while for the DBS electrode all contacts were inserted in the HH. After recording we decided to explore the part of the HH closest to the disconnection region to understand if there was a gradient in epileptic activity. Scalp EEG was performed in all patients (referential montage, bimastoid reference). Anesthesiology protocol Intravenous general anesthesia was performed with propofol (2.5 mg/kg as bolus and 8–9 mg/kg as maintenance dose) and remifentanil (0.1–0.5 lg/kg/min), with a single shot of vecuronium bromide (0.08 mg/kg) for orotracheal intubation. During surgery all drugs were titrated to
6 months
2 months
2 months
2
3
4
Outcome after the second operation
Age at second operation
Gelastic (3/ day) Focal (daily)
Gelastic (daily) Focal and SG (rare)
Facial grimace (3–4/day) Gelastic and focal (3/week)
Gelastic and focal (10/ month) Focal SG (the first one)
Seizure type and frequency
Diffuse rhythmic theta activity
Diffuse rhythmic theta activity
R temporal sharp waves
Bilateral C–P theta-delta waves, isolated or clustered spike-waves
Bilateral temporal epileptiform abnormalities (R [ L)
Temporal epileptiform abnormalities (L [ R)
L temporal rhythmic thetadelta activity
Diffuse lowvoltage fast activity
Interictal EEG
Ictal EEG (gelastic seizures)
VPA, LEVa, CBZa
CBZ, VPAa, CLBa
TPM, PB, VPA, LTG, CBZa
PB, CBZ, LEVa, VGBa
Therapy
Unchanged
Unchanged
Hyperactivity, attention deficit
Slight psychomotor delay
Unchanged
Unchanged
Post-op neurol status
Normal
Normal
Pre-op neurol status
Normal
Precocious puberty
Normal
Normal
Endocr pre-op
Unchanged
Triptorelin
Unchanged
Unchanged
Endocr post-op
2 year 11 months 4 years 3 monthsb
7 years 2 months 7 years 7 monthsb
17 years
31 years
Age at surgery
theta-beta activity intermingled with infrequent slow waves
Right temporal and central slow waves
Slow wave superimposed by fast activity
Spikes over the left temporal lobe
Intraoperative recording (scalp)
High-amplitude sharp waves, intermingled with numerous spikes
Continuous fast activity with very few medium amplitude sharp waves
Rare paroxysms of sharp waves with few spikes, on a background of beta activity
Synchronous epileptiform abnormalities on the L side
Intraoperative recording (lesion)
2) Seizure freec
1) 1/die
16
3
1) \ 50 % 2) Seizure freec
12
18
F–U (months)
Seizure free
Unchanged
Outcome
Pt patient, Del Cl Delalande classification, d day, w week, m months ,y years, SG secondarily generalized, L left, R right, C-P centro-parietal, F-C fronto-central, C-T centro-temporal, PB phenobarbital, CBZ carbamazepine, LEV levetiracetam, VGV vigabatrin, TPM topiramate, VPA valproic acid, LTG lamotrigine, CLB clobazam, F-U follow-up, Pre-op pre-operative, Post-op post-operative, Endocr endocrinologic
c
b
2
4
2
4
Del Cl
Current treatment
13 years
1
a
Age onset
Pt
Table 1 Clinical, neurophysiologic findings, and outcome in 4 patients with an acute intralesional recording of hypothalamic hamartoma
Acta Neurol Belg
123
Acta Neurol Belg Fig. 1 Pre-operative brain MR of patients #2, #3, #4. Scalp and contemporary intra-hamartoma EEG recordings. Patients were anesthetized during the operation. a Brain MR in patient #2: coronal IR image showing type 2 hypothalamic hamartoma. b Brain MR in patient #3: coronal T2-weighted image showing type 4 hypothalamic hamartoma. c Brain MR in patient #4: coronal T2-weighted image showing type 2 hypothalamic hamartoma. d Recording in patient #2: scalp EEG shows slow wave superimposed by fast activity, intra-hamartoma recording shows rare paroxysms of sharp waves with few spikes, on a background of beta activity. e Recording in patient #3: scalp EEG shows intermittent right central and temporal high-amplitude slow waves intermingled with low voltage theta waves, intrahamartoma recording shows continuous fast activity with very few medium amplitude sharp waves, in brief trains. f Recording in patient #4: scalp EEG shows theta-beta activity intermingled with infrequent slow waves, intra-hamartoma recording shows recurrent highamplitude sharp waves, intermingled with numerous spikes, on a very low amplitude background activity
maintain the value of the bispectral index monitoring (BIS) between 30 and 60. A BIS value of 60 was reached (by reduction of propofol and remifentanil infusion), to allow for intraoperative EEG monitoring.
Results Table 1 summarizes all clinical and neurophysiological data of the 4 patients in which intraoperative recordings were performed. Figure 1 shows pre-operative brain MR (A–C) and intraoperative EEG recording (D–F) in three out of four cases (Patients #2, #3, #4).
123
Case report #1 This is a 32-year-old right-handed woman. A detailed description of the clinical findings has already been reported [16]. Epilepsy onset was at 13-year-old. She presented with gelastic and focal seizures, characterized by staring and automatisms sometimes followed by secondary generalization. Seizure duration ranged between 20 and 60 s and occurred about 10 times per month. AEDs were ineffective. Interictal EEG revealed epileptiform abnormalities bilateral over temporal lobes, prevalent on the left side and increased during sleep. A complex focal seizure was recorded: the ictal discharge was characterized by
Acta Neurol Belg
accumulating in trains of variable duration (3–10 s). Soon after surgery she was seizure free; follow-up duration is 3 years. Case report #3
Fig. 2 Endoscopic direct recording with the deep electrode in correspondence of the hypothalamic hamartoma during the disconnection procedure of patient #3
theta-delta rhythmic recruiting activity over the left temporal lobe. MR revealed a HH. Cognitive development was normal. A complete hormonal assessment was normal (Fig. 2). Patient refused resective surgery and a bilateral DBS electrodes implantation was performed. Intraoperative scalp-EEG recording showed synchronous interictal epileptiform abnormalities over the left temporal region and inside the lesion (on the left side). The maximum amplitude of the spikes was observed on the distal contacts of the left DBS electrode (contacts 0-1-2) and on the F7 and T3 scalp electrodes. No interictal epileptic abnormalities were recorded from the right side of the hamartoma. This treatment determined a slight reduction of seizure frequency. Case report #2 The patient is an 18-year-old right-handed girl, whose seizures started at the age of 6 months. Seizures semiology was characterized by left facial grimace, accumulated in 3–4 clusters per day, mostly during sleep. Seizure duration was of 4–5 s. At the age of 4 years she started to experience gelastic seizures in addition, sometimes followed by left hemiconvulsion. Seizures occurred 3 times per week and duration ranged from 1 to 2 min. Pharmacological treatment resulted ineffective. Interictal EEG revealed epileptiform abnormalities over bilateral temporal regions, prevalent on the right side; ictal EEG of gelastic seizures showed diffuse low-voltage fast activity, without a clear epileptiform discharge. Brain MR revealed a class 2 HH, strongly connected mostly to the right hypothalamic structures. The patient underwent right-side HH stereoendoscopic disconnection. The scalp intraoperative recordings revealed slow waves superimposed by fast activity, while over the intralesional electrode paroxysms of sharp waves and spikes of brief duration, intermingled with low-voltage activity were evident. The activity tended to be persistent throughout the recording, often
This is an 8-year-old boy, with normal psychomotor development. Since the age of 4 months he presented with episodes of facial grimace of brief duration. These episodes were almost daily and were not recognized as seizures by parents till the age of 7 years, when an episode characterized by left arm clonic jerks, left head deviation and sialorrhea occurred. Brain MR revealed a huge class 4 HH, involving both sides of hypothalamus. Interictal EEG showed a bilateral centro-parietal theta-delta activity and isolated or clustered spikes and waves complexes on the same regions. Ictal EEG revealed medium–low voltage diffuse fast activity. Mild behavioral disturbances were also reported. Precocious puberty was also evident, treated with Triptoreline. Endoscopic disconnection of HH was performed in two steps at the age of 7 years and 2 months and after 5 months. During the first procedure acute intralesional recording was performed: interictal scalp intraoperative EEG showed right central and temporal highamplitude slow waves intermingled with low voltage theta waves, while the intralesional electrode revealed continuous fast activity with very few medium amplitude sharp waves, in brief trains. This activity was almost continuous and was more evident over the first three contacts (deeper); a gradient between internal and external contacts was documented, suggesting that the internal contacts were located in the most epileptogenic area of the HH. After the second disconnection the patient was seizure free. Case report #4 This is a 5 year-old boy, born after uneventful pregnancy and delivery. Neonatal period was regular; he walked at the age of 18 months. At the age of 2 months he started to present with episodes characterized by loss of contact by right eye deviation and diffuse hypertonus. Since the age of 6 months parents reported gelastic seizures characterized by loss of contact and laughing. Gelastic seizures recurred almost daily since the onset. Interictal EEG showed right temporal slow waves and ictal EEG revealed a diffuse low voltage fast activity more prominent over the same region. Neuropsychological evaluation, performed with the Griffith’s scale, revealed a total IQ of 82. He was treated with valproate, carbamazepine and levetiracetam, without effects. Brain MR showed a class 2 HH mainly involving the right hypothalamus. He underwent a stereo-endoscopic right disconnection at the age of 2 years and 11 months.
123
Acta Neurol Belg
The scalp intraoperative recording showed theta-beta activity intermingled with infrequent slow waves. The intralesional electrode documented high-amplitude sharp waves, intermingled with several spikes. Clear-cut runs of poly-spikes or trains of low-voltage fast activity were not documented. The patient, after surgery, presented daily gelastic seizures. The following intervention, 18 months later (4 years and 3 months), allowed for the disconnection of the left side of the hamartoma resulting in seizure cessation over a 20-month follow-up.
Discussion HH are lesions intrinsically epileptogenic, from which both gelastic and focal seizures might originate. This was firstly reported by Munari et al. in 1995 by means of EEG recordings with depth electrodes implanted in the hamartoma and later confirmed by the reproduction of typical gelastic seizures by HH stimulation [13, 17]. To date, intralesional EEG recording during the HH disconnection procedure was performed in few cases [16, 18]. Stereotactic intracerebral EEG recording was preferred to understand the neurophysiological pathway of seizures arising from the HH and spreading to adjacent cortex giving rise to different types of focal seizures. Moreover, the stereotactic intracerebral EEG recording allowed the reproduction of gelastic or dacrystic seizures by stimulation procedures. The clinical characteristics in our cases were similar to those reported for drug-resistant epilepsy associated to HH. In one case we also documented central precocious puberty without other neurological symptoms. We recorded intralesional electrical activity during stereo-endoscopic disconnection in three cases and during DBS implantation in one. We also correlated it with the simultaneous scalp-EEG recording. While HH disconnection resulted in complete seizure remission, DBS was ineffective. As previously reported, ictal scalp EEG during gelastic seizures showed a diffuse low-voltage fast activity associated with decreasing or disappearance of interictal epileptiform activity. This finding had non-localizing value in the majority of cases. On the other hand, focal seizures associated to HH, in some cases have shown localizing ictal patterns over the temporal or frontal regions, more often on the same side of HH attachment [12]. Interictal scalp-EEG recordings in our cases mostly showed diffuse abnormalities and rarely focal, temporal or frontal, slow or epileptiform abnormalities as already pointed out [13, 16]. Therefore, both the ictal and the
123
interictal scalp EEG, have poor utility in patients with HH [12]. The ictal discharge arising from the HH had been already documented with stereo-EEG recordings [14, 16, 17]. Acute intraoperative recordings in our cases confirmed the presence of epileptiform abnormalities intermingled with low-voltage activity, mostly on the same side of the HH attachment. Paroxysmal activity recorded inside the HH was always evident. It was difficult to relate the paroxysmal activity recorded inside the hamartoma with the scalp EEG, because in some cases epileptic activity does not spread to the cortex but remains confined inside the hamartoma. This paper presents preliminary data on the use of neurophysiologic procedures useful for a guided disconnection or for small resections. It might be useful to eventually suggest disconnections associated with small endoscopic resections keeping a low rate of morbidity, and to understand the value of interictal scalp-EEG activity. In our experience, HH stereo-endoscopic disconnection is a safe procedure, not burdened by a high rate of complications, and for this reason is reproducible. Two out of three cases had a second disconnection due to the bilateral adhesion of HH to the third ventricle. Mapping of HH epileptogenic activity could be useful to confirm the usefulness of disconnection procedure. This should consider on-site recording from the HH and if abnormalities are detected safely proceed to disconnection of the HH. Conflict of interest of interest.
The authors declare that they have no conflict
References 1. Coons SW, Rekate HL, Prenger EC et al (2007) The histopathology of hypothalamic hamartomas: study of 57 cases. J Neuropathol Exp Neurol 66:131–141 2. Oehl B, Brandt A, Fauser S et al (2010) Semiologic aspects of epileptic seizures in 31 patients with hypothalamic hamartoma. Epilepsia 51:2116–2123 3. Parvizi J, Le S, Foster BL, Bourgeois B et al (2011) Gelastic epilepsy and hypothalamic hamartomas: neuroanatomical analysis of brain lesions in 100 patients. Brain 134:2960–2968 4. 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 43:61–68 5. Fohlen M, Lellouch A, Delalande O (2003) Hypothalamic hamartoma with refractory epilepsy: surgical procedures and results in 18 patients. Epileptic Disord 5:267–273 6. Dorfmu¨ller G, Fohlen M, Bulteau C et al (2008) Surgical disconnection of hypothalamic hamartomas. Neurochirurgie 54:315–319 7. Harvey AS, Freeman JL (2007) Epilepsy in hypothalamic hamartoma: clinical and EEG features. Semin Pediatr Neurol 14:60–64
Acta Neurol Belg 8. Mathieu D, Deacon C, Pinard CA et al (2010) Gamma knife surgery for hypothalamic hamartomas causing refractory epilepsy: preliminary results from a prospective observational study. J Neurosurg 113:215–221 9. Schulze-Bonhage A, Homberg V, Trippel M et al (2004) Interstitial radiosurgery in the treatment of gelastic epilepsy due to hypothalamic hamartomas. Neurology 62:644–647 10. Re´gis J, Scavarda D, Tamura M et al (2007) Gamma knife surgery for epilepsy related to hypothalamic hamartomas. Semin Pediatr Neurol 14:73–79 11. Arita K, Ikawa F, Kurisu K et al (1999) The relationship between magnetic resonance imaging findings and clinical manifestations of hypothalamic hamartoma. J Neurosurg 91:212–220 12. Troester M, Haine-Schlagel R, Ng YT et al (2011) EEG and video-EEG seizure monitoring has limited utility in patients with hypothalamic hamartoma and epilepsy. Epilepsia 52:1137–1143 13. Kahane P, Ryvlin P, Hoffmann D et al (2003) From hypothalamic hamartoma to cortex: what can be learnt from depth recordings and stimulation? Epileptic Disord 5:205–217
14. Munari C, Kahane P, Francione S et al (1995) Role of the hypothalamic hamartoma in the genesis of gelastic fits (a videostereo-EEG study). Electroencephalogr Clin Neurophysiol 95:154–160 15. Blume WT, Lu¨ders HO, Mizrahi E et al (2001) Glossary of descriptive terminology for ictal semiology: report of the ILAE task force on classification and terminology. Epilepsia 42:1212–1218 16. Marras CE, Rizzi M, Villani F et al (2011) Deep brain stimulation for the treatment of drug-refractory epilepsy in a patient with a hypothalamic hamartoma. Case report. Neurosurg Focus 30:E4 17. Kuzniecky R, Guthrie B, Mountz J et al (1997) Intrinsic epileptogenesis of hypothalamic hamartomas in gelastic epilepsy. Ann Neurol 42:60–67 18. Roberts CM, Thompson EM, Selden NR (2011) Transendoscopic intraoperative recording of gelastic seizures from a hypothalamic hamartoma. Pediatr Neurosurg 47:147–151
123
ORIGINAL ARTICLE
Acute intralesional recording in hypothalamic hamartoma: description of 4 cases Nicola Specchio • Michele Rizzi • Marina Trivisano • Lucia Fusco • Erica Rebessi • Simona Cappelletti • Luca De Palma • Flavio Villani • Alessandra Savioli • Alessandro De Benedictis • Carlo Efisio Marras • Federico Vigevano • Olivier Delalande
Received: 21 July 2014 / Accepted: 30 September 2014 Ó Belgian Neurological Society 2014
Abstract Hypothalamic hamartomas (HHs) are intrinsically epileptogenic lesions associated to medically intractable focal epilepsy mainly characterized by gelastic and focal seizures. Intralesional recording with deep electrodes has documented the presence of ictal discharge arising from inside the lesion. Nevertheless interictal and ictal scalp EEG is poorly informative and non-localizing in a great deal of cases. HH disconnection leads to seizure remission in most cases. To describe the intralesional EEG recordings and to compare them with concomitant scalp EEG and with previous cases reported in literature. We reviewed the medical records of 17 children affected by drug-resistant focal epilepsy associated to HH. We recorded intralesional electrical activity during stereo-endoscopic disconnection in three cases and during deep brain stimulation implantation in one. We also correlated it with
Electronic supplementary material The online version of this article (doi:10.1007/s13760-014-0374-z) contains supplementary material, which is available to authorized users. N. Specchio (&) L. Fusco L. De Palma F. Vigevano Division of Neurology, Department of Neuroscience, Bambino Gesu` Children’s Hospital IRCCS, P.za S. Onofrio, 4 00165 Rome, Italy e-mail: [email protected] M. Rizzi Division of Neurosurgery, IRCCS Neurological Institute ‘‘C.Besta’’, Milan, Italy M. Trivisano Clinic of Nervous System Diseases, University of Foggia, Foggia, Italy E. Rebessi A. De Benedictis C. E. Marras O. Delalande Division of Neurosurgery, Department of Neuroscience, Bambino Gesu` Children’s Hospital IRCCS, Rome, Italy
the simultaneous scalp-EEG recording. Acute intralesional recordings in our cases confirmed the presence of epileptiform abnormalities intermingled with low-voltage activity, mostly on the same side of the HH attachment. Paroxysmal activity recorded inside the HH was always evident. Mapping of HH epileptogenic activity could be useful to confirm the usefulness of disconnection procedure. This should consider on-site recording from the HH and if abnormalities are detected safely proceed to disconnection of the HH. Keywords Hypothalamic hamartoma Gelastic seizures Depth electrodes Invasive monitoring Epilepsy surgery
Introduction Hypothalamic hamartomas (HHs) are well-recognized congenital malformations of the brain, defined as benign, tumor-like nodules of disorganized but mature cells mostly S. Cappelletti Psychology Unit, Department of Neuroscience, Bambino Gesu` Children’s Hospital IRCCS, Rome, Italy F. Villani Division of Clinical Epileptology and Experimental Neurophysiology, Fondazione Istituto Neurologico ‘‘Carlo Besta’’, Milan, Italy A. Savioli Division of Anaesthesia and Emergency Room, Bambino Gesu` Children’s Hospital IRCCS, Rome, Italy O. Delalande Service de Neurochirurgie Pe´diatrique, Fondation Ophtalmologique Adolphe de Rothschild, Paris, France
123
Acta Neurol Belg
composed of a combination of neuronal or ganglion cells, glial cells and blood vessels [1]. HH can be responsible for neurological and endocrine disturbances. HH represent a cause of medically intractable focal epilepsy. Antiepileptic drugs (AEDs) are poorly effective in patients with HH [2, 3] therefore early surgical treatment should be considered in every case [4, 5]. Two different surgical approaches, resection or disconnection, can lead to remission of the attacks, even after AED withdrawal [6, 7]. There is also evidence that radiofrequency and Gamma knife radiosurgery may lead to a complete seizure control with limited morbidity, when the entire lesion can be targeted [8–11]. HH produces stereotyped epileptic seizures characterized by paroxysmal laughing or crying (gelastic or dacrystic seizures), other than focal seizures with or without motor manifestations [2]. Seizures onset is usually in the first months of life and frequency is variable from daily to weekly recurrence. Interictal and ictal scalp EEG is usually poorly informative in most cases: focal or diffuse spike and wave complexes have been recorded in some cases in some cases, focal theta and delta activity could also be associated with epileptiform abnormalities [12]. Ictal EEG usually shows diffuse flattening or focal theta rhythmic activity. Interictal abnormalities and ictal changes usually predominate over the side ipsilateral to the hamartoma [13]. Intralesional recording with an electrode implanted in the hamartoma showed the presence of an ictal discharge arising inside the lesion associated to the clinical pattern [13, 14]. Within a collaborative epilepsy surgery program we studied 17 patients with drug-resistant epilepsy associated to HH. Acute intralesional recording of the interictal activity arising inside the HH was performed during deep brain stimulation (DBS) electrode implantation, and endoscopic disconnections. The aim of this report is to describe the results of intralesional recordings, to compare them with concurrent scalp-EEG recordings and with previous cases reported in literature.
Methods We have reviewed the medical records of 17 children affected by drug-resistant focal epilepsy associated to HH, referred between January 1990 and December 2011 to the Neurology Division of the Bambino Gesu` Children’s Hospital in Rome and ‘‘Carlo Besta’’ Neurological Institute in Milan: both places receive tertiary patients for specialist consultations. All clinical notes were reviewed to evaluate the age at onset, the seizure type at onset and during the follow-up,
123
the treatment and the epilepsy outcome. Seizures were defined according to the Glossary of Descriptive Terminology for Ictal Semiology [15]. HHs were defined according to the classification proposed by Delalande [4]. Mean patient age at seizure onset was 4,67 ±3.9 years (range 1 day–13 years). At onset 14 patients presented with gelastic seizures, 1 patient with facial grimace, 1 patient with hemifacial spasms associated to ocular blinking, and 1 with ictal syncope. During the follow-up other type of seizures appeared: generalized tonic–clonic (3 pts), tonic (1 pt), atonic (1 pt), clonic (1 pt), and dacrystic (1 pt). In 10 out of 17 cases a surgical approach was performed: 4 of them underwent a one stage stereo-endoscopic disconnection, 3 cases a double stage stereo-endoscopic disconnection, and 1 case required a multistage stage stereoendoscopic disconnection, 1 patient performed a DBS protocol, and 1 a surgical resection. Moreover, 2 patients were treated with radiotherapy, and radiosurgery. Following Delalande classification [4], HHs were mostly of type 2 (8 out of 17), followed by type 3 (6 out of 17) and type 4 (3 out of 17). No cases were classified as type 1. See Supplementary Table 1. In four patients an intraoperative endoscopic intralesional recording was performed. In these cases we also collected data regarding pre- and post-operative neurological, neuropsychological and endocrinological evaluations, EEG findings including interictal and ictal long-term monitoring; moreover, a detailed treatment history was reported. Clinical and neurophysiological data of patient 1 have already been reported by Marras et al. [16]. In 3 cases intraoperative stereo EEG was carried out using foramen ovale electrodes (Dixi Medical ACS-798S, 5 contacts), and DBS leads in 1 case (Medtronic 3389, 4 contacts: 1.27 mm diameter, 1.5 mm long, spaced by 0.5 mm). Foramen ovale electrodes were introduced in the core of the hamartomas and activity was recorded for at least 10 min. For the foramen ovale electrodes only the first 4 contacts were inserted and activity was recorded in referential montage, while for the DBS electrode all contacts were inserted in the HH. After recording we decided to explore the part of the HH closest to the disconnection region to understand if there was a gradient in epileptic activity. Scalp EEG was performed in all patients (referential montage, bimastoid reference). Anesthesiology protocol Intravenous general anesthesia was performed with propofol (2.5 mg/kg as bolus and 8–9 mg/kg as maintenance dose) and remifentanil (0.1–0.5 lg/kg/min), with a single shot of vecuronium bromide (0.08 mg/kg) for orotracheal intubation. During surgery all drugs were titrated to
6 months
2 months
2 months
2
3
4
Outcome after the second operation
Age at second operation
Gelastic (3/ day) Focal (daily)
Gelastic (daily) Focal and SG (rare)
Facial grimace (3–4/day) Gelastic and focal (3/week)
Gelastic and focal (10/ month) Focal SG (the first one)
Seizure type and frequency
Diffuse rhythmic theta activity
Diffuse rhythmic theta activity
R temporal sharp waves
Bilateral C–P theta-delta waves, isolated or clustered spike-waves
Bilateral temporal epileptiform abnormalities (R [ L)
Temporal epileptiform abnormalities (L [ R)
L temporal rhythmic thetadelta activity
Diffuse lowvoltage fast activity
Interictal EEG
Ictal EEG (gelastic seizures)
VPA, LEVa, CBZa
CBZ, VPAa, CLBa
TPM, PB, VPA, LTG, CBZa
PB, CBZ, LEVa, VGBa
Therapy
Unchanged
Unchanged
Hyperactivity, attention deficit
Slight psychomotor delay
Unchanged
Unchanged
Post-op neurol status
Normal
Normal
Pre-op neurol status
Normal
Precocious puberty
Normal
Normal
Endocr pre-op
Unchanged
Triptorelin
Unchanged
Unchanged
Endocr post-op
2 year 11 months 4 years 3 monthsb
7 years 2 months 7 years 7 monthsb
17 years
31 years
Age at surgery
theta-beta activity intermingled with infrequent slow waves
Right temporal and central slow waves
Slow wave superimposed by fast activity
Spikes over the left temporal lobe
Intraoperative recording (scalp)
High-amplitude sharp waves, intermingled with numerous spikes
Continuous fast activity with very few medium amplitude sharp waves
Rare paroxysms of sharp waves with few spikes, on a background of beta activity
Synchronous epileptiform abnormalities on the L side
Intraoperative recording (lesion)
2) Seizure freec
1) 1/die
16
3
1) \ 50 % 2) Seizure freec
12
18
F–U (months)
Seizure free
Unchanged
Outcome
Pt patient, Del Cl Delalande classification, d day, w week, m months ,y years, SG secondarily generalized, L left, R right, C-P centro-parietal, F-C fronto-central, C-T centro-temporal, PB phenobarbital, CBZ carbamazepine, LEV levetiracetam, VGV vigabatrin, TPM topiramate, VPA valproic acid, LTG lamotrigine, CLB clobazam, F-U follow-up, Pre-op pre-operative, Post-op post-operative, Endocr endocrinologic
c
b
2
4
2
4
Del Cl
Current treatment
13 years
1
a
Age onset
Pt
Table 1 Clinical, neurophysiologic findings, and outcome in 4 patients with an acute intralesional recording of hypothalamic hamartoma
Acta Neurol Belg
123
Acta Neurol Belg Fig. 1 Pre-operative brain MR of patients #2, #3, #4. Scalp and contemporary intra-hamartoma EEG recordings. Patients were anesthetized during the operation. a Brain MR in patient #2: coronal IR image showing type 2 hypothalamic hamartoma. b Brain MR in patient #3: coronal T2-weighted image showing type 4 hypothalamic hamartoma. c Brain MR in patient #4: coronal T2-weighted image showing type 2 hypothalamic hamartoma. d Recording in patient #2: scalp EEG shows slow wave superimposed by fast activity, intra-hamartoma recording shows rare paroxysms of sharp waves with few spikes, on a background of beta activity. e Recording in patient #3: scalp EEG shows intermittent right central and temporal high-amplitude slow waves intermingled with low voltage theta waves, intrahamartoma recording shows continuous fast activity with very few medium amplitude sharp waves, in brief trains. f Recording in patient #4: scalp EEG shows theta-beta activity intermingled with infrequent slow waves, intra-hamartoma recording shows recurrent highamplitude sharp waves, intermingled with numerous spikes, on a very low amplitude background activity
maintain the value of the bispectral index monitoring (BIS) between 30 and 60. A BIS value of 60 was reached (by reduction of propofol and remifentanil infusion), to allow for intraoperative EEG monitoring.
Results Table 1 summarizes all clinical and neurophysiological data of the 4 patients in which intraoperative recordings were performed. Figure 1 shows pre-operative brain MR (A–C) and intraoperative EEG recording (D–F) in three out of four cases (Patients #2, #3, #4).
123
Case report #1 This is a 32-year-old right-handed woman. A detailed description of the clinical findings has already been reported [16]. Epilepsy onset was at 13-year-old. She presented with gelastic and focal seizures, characterized by staring and automatisms sometimes followed by secondary generalization. Seizure duration ranged between 20 and 60 s and occurred about 10 times per month. AEDs were ineffective. Interictal EEG revealed epileptiform abnormalities bilateral over temporal lobes, prevalent on the left side and increased during sleep. A complex focal seizure was recorded: the ictal discharge was characterized by
Acta Neurol Belg
accumulating in trains of variable duration (3–10 s). Soon after surgery she was seizure free; follow-up duration is 3 years. Case report #3
Fig. 2 Endoscopic direct recording with the deep electrode in correspondence of the hypothalamic hamartoma during the disconnection procedure of patient #3
theta-delta rhythmic recruiting activity over the left temporal lobe. MR revealed a HH. Cognitive development was normal. A complete hormonal assessment was normal (Fig. 2). Patient refused resective surgery and a bilateral DBS electrodes implantation was performed. Intraoperative scalp-EEG recording showed synchronous interictal epileptiform abnormalities over the left temporal region and inside the lesion (on the left side). The maximum amplitude of the spikes was observed on the distal contacts of the left DBS electrode (contacts 0-1-2) and on the F7 and T3 scalp electrodes. No interictal epileptic abnormalities were recorded from the right side of the hamartoma. This treatment determined a slight reduction of seizure frequency. Case report #2 The patient is an 18-year-old right-handed girl, whose seizures started at the age of 6 months. Seizures semiology was characterized by left facial grimace, accumulated in 3–4 clusters per day, mostly during sleep. Seizure duration was of 4–5 s. At the age of 4 years she started to experience gelastic seizures in addition, sometimes followed by left hemiconvulsion. Seizures occurred 3 times per week and duration ranged from 1 to 2 min. Pharmacological treatment resulted ineffective. Interictal EEG revealed epileptiform abnormalities over bilateral temporal regions, prevalent on the right side; ictal EEG of gelastic seizures showed diffuse low-voltage fast activity, without a clear epileptiform discharge. Brain MR revealed a class 2 HH, strongly connected mostly to the right hypothalamic structures. The patient underwent right-side HH stereoendoscopic disconnection. The scalp intraoperative recordings revealed slow waves superimposed by fast activity, while over the intralesional electrode paroxysms of sharp waves and spikes of brief duration, intermingled with low-voltage activity were evident. The activity tended to be persistent throughout the recording, often
This is an 8-year-old boy, with normal psychomotor development. Since the age of 4 months he presented with episodes of facial grimace of brief duration. These episodes were almost daily and were not recognized as seizures by parents till the age of 7 years, when an episode characterized by left arm clonic jerks, left head deviation and sialorrhea occurred. Brain MR revealed a huge class 4 HH, involving both sides of hypothalamus. Interictal EEG showed a bilateral centro-parietal theta-delta activity and isolated or clustered spikes and waves complexes on the same regions. Ictal EEG revealed medium–low voltage diffuse fast activity. Mild behavioral disturbances were also reported. Precocious puberty was also evident, treated with Triptoreline. Endoscopic disconnection of HH was performed in two steps at the age of 7 years and 2 months and after 5 months. During the first procedure acute intralesional recording was performed: interictal scalp intraoperative EEG showed right central and temporal highamplitude slow waves intermingled with low voltage theta waves, while the intralesional electrode revealed continuous fast activity with very few medium amplitude sharp waves, in brief trains. This activity was almost continuous and was more evident over the first three contacts (deeper); a gradient between internal and external contacts was documented, suggesting that the internal contacts were located in the most epileptogenic area of the HH. After the second disconnection the patient was seizure free. Case report #4 This is a 5 year-old boy, born after uneventful pregnancy and delivery. Neonatal period was regular; he walked at the age of 18 months. At the age of 2 months he started to present with episodes characterized by loss of contact by right eye deviation and diffuse hypertonus. Since the age of 6 months parents reported gelastic seizures characterized by loss of contact and laughing. Gelastic seizures recurred almost daily since the onset. Interictal EEG showed right temporal slow waves and ictal EEG revealed a diffuse low voltage fast activity more prominent over the same region. Neuropsychological evaluation, performed with the Griffith’s scale, revealed a total IQ of 82. He was treated with valproate, carbamazepine and levetiracetam, without effects. Brain MR showed a class 2 HH mainly involving the right hypothalamus. He underwent a stereo-endoscopic right disconnection at the age of 2 years and 11 months.
123
Acta Neurol Belg
The scalp intraoperative recording showed theta-beta activity intermingled with infrequent slow waves. The intralesional electrode documented high-amplitude sharp waves, intermingled with several spikes. Clear-cut runs of poly-spikes or trains of low-voltage fast activity were not documented. The patient, after surgery, presented daily gelastic seizures. The following intervention, 18 months later (4 years and 3 months), allowed for the disconnection of the left side of the hamartoma resulting in seizure cessation over a 20-month follow-up.
Discussion HH are lesions intrinsically epileptogenic, from which both gelastic and focal seizures might originate. This was firstly reported by Munari et al. in 1995 by means of EEG recordings with depth electrodes implanted in the hamartoma and later confirmed by the reproduction of typical gelastic seizures by HH stimulation [13, 17]. To date, intralesional EEG recording during the HH disconnection procedure was performed in few cases [16, 18]. Stereotactic intracerebral EEG recording was preferred to understand the neurophysiological pathway of seizures arising from the HH and spreading to adjacent cortex giving rise to different types of focal seizures. Moreover, the stereotactic intracerebral EEG recording allowed the reproduction of gelastic or dacrystic seizures by stimulation procedures. The clinical characteristics in our cases were similar to those reported for drug-resistant epilepsy associated to HH. In one case we also documented central precocious puberty without other neurological symptoms. We recorded intralesional electrical activity during stereo-endoscopic disconnection in three cases and during DBS implantation in one. We also correlated it with the simultaneous scalp-EEG recording. While HH disconnection resulted in complete seizure remission, DBS was ineffective. As previously reported, ictal scalp EEG during gelastic seizures showed a diffuse low-voltage fast activity associated with decreasing or disappearance of interictal epileptiform activity. This finding had non-localizing value in the majority of cases. On the other hand, focal seizures associated to HH, in some cases have shown localizing ictal patterns over the temporal or frontal regions, more often on the same side of HH attachment [12]. Interictal scalp-EEG recordings in our cases mostly showed diffuse abnormalities and rarely focal, temporal or frontal, slow or epileptiform abnormalities as already pointed out [13, 16]. Therefore, both the ictal and the
123
interictal scalp EEG, have poor utility in patients with HH [12]. The ictal discharge arising from the HH had been already documented with stereo-EEG recordings [14, 16, 17]. Acute intraoperative recordings in our cases confirmed the presence of epileptiform abnormalities intermingled with low-voltage activity, mostly on the same side of the HH attachment. Paroxysmal activity recorded inside the HH was always evident. It was difficult to relate the paroxysmal activity recorded inside the hamartoma with the scalp EEG, because in some cases epileptic activity does not spread to the cortex but remains confined inside the hamartoma. This paper presents preliminary data on the use of neurophysiologic procedures useful for a guided disconnection or for small resections. It might be useful to eventually suggest disconnections associated with small endoscopic resections keeping a low rate of morbidity, and to understand the value of interictal scalp-EEG activity. In our experience, HH stereo-endoscopic disconnection is a safe procedure, not burdened by a high rate of complications, and for this reason is reproducible. Two out of three cases had a second disconnection due to the bilateral adhesion of HH to the third ventricle. Mapping of HH epileptogenic activity could be useful to confirm the usefulness of disconnection procedure. This should consider on-site recording from the HH and if abnormalities are detected safely proceed to disconnection of the HH. Conflict of interest of interest.
The authors declare that they have no conflict
References 1. Coons SW, Rekate HL, Prenger EC et al (2007) The histopathology of hypothalamic hamartomas: study of 57 cases. J Neuropathol Exp Neurol 66:131–141 2. Oehl B, Brandt A, Fauser S et al (2010) Semiologic aspects of epileptic seizures in 31 patients with hypothalamic hamartoma. Epilepsia 51:2116–2123 3. Parvizi J, Le S, Foster BL, Bourgeois B et al (2011) Gelastic epilepsy and hypothalamic hamartomas: neuroanatomical analysis of brain lesions in 100 patients. Brain 134:2960–2968 4. 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 43:61–68 5. Fohlen M, Lellouch A, Delalande O (2003) Hypothalamic hamartoma with refractory epilepsy: surgical procedures and results in 18 patients. Epileptic Disord 5:267–273 6. Dorfmu¨ller G, Fohlen M, Bulteau C et al (2008) Surgical disconnection of hypothalamic hamartomas. Neurochirurgie 54:315–319 7. Harvey AS, Freeman JL (2007) Epilepsy in hypothalamic hamartoma: clinical and EEG features. Semin Pediatr Neurol 14:60–64
Acta Neurol Belg 8. Mathieu D, Deacon C, Pinard CA et al (2010) Gamma knife surgery for hypothalamic hamartomas causing refractory epilepsy: preliminary results from a prospective observational study. J Neurosurg 113:215–221 9. Schulze-Bonhage A, Homberg V, Trippel M et al (2004) Interstitial radiosurgery in the treatment of gelastic epilepsy due to hypothalamic hamartomas. Neurology 62:644–647 10. Re´gis J, Scavarda D, Tamura M et al (2007) Gamma knife surgery for epilepsy related to hypothalamic hamartomas. Semin Pediatr Neurol 14:73–79 11. Arita K, Ikawa F, Kurisu K et al (1999) The relationship between magnetic resonance imaging findings and clinical manifestations of hypothalamic hamartoma. J Neurosurg 91:212–220 12. Troester M, Haine-Schlagel R, Ng YT et al (2011) EEG and video-EEG seizure monitoring has limited utility in patients with hypothalamic hamartoma and epilepsy. Epilepsia 52:1137–1143 13. Kahane P, Ryvlin P, Hoffmann D et al (2003) From hypothalamic hamartoma to cortex: what can be learnt from depth recordings and stimulation? Epileptic Disord 5:205–217
14. Munari C, Kahane P, Francione S et al (1995) Role of the hypothalamic hamartoma in the genesis of gelastic fits (a videostereo-EEG study). Electroencephalogr Clin Neurophysiol 95:154–160 15. Blume WT, Lu¨ders HO, Mizrahi E et al (2001) Glossary of descriptive terminology for ictal semiology: report of the ILAE task force on classification and terminology. Epilepsia 42:1212–1218 16. Marras CE, Rizzi M, Villani F et al (2011) Deep brain stimulation for the treatment of drug-refractory epilepsy in a patient with a hypothalamic hamartoma. Case report. Neurosurg Focus 30:E4 17. Kuzniecky R, Guthrie B, Mountz J et al (1997) Intrinsic epileptogenesis of hypothalamic hamartomas in gelastic epilepsy. Ann Neurol 42:60–67 18. Roberts CM, Thompson EM, Selden NR (2011) Transendoscopic intraoperative recording of gelastic seizures from a hypothalamic hamartoma. Pediatr Neurosurg 47:147–151
123
