SEARCH

SEARCH BY CITATION

Keywords:

  • Cerebellar ganglioglioma;
  • Epileptic seizure;
  • Infancy

Abstract

  1. Top of page
  2. Abstract
  3. CASE REPORT
  4. DISCUSSION
  5. REFERENCES

Summary: The cerebellum is known to have an inhibitory effect on seizures. Nevertheless, cerebellar dysplastic lesions can be epileptogenic. A 4-month-old infant had paroxysmal facial contractions; tachypnea and nystagmoid eyeball and tremulous movements were occasionally combined. These evolved to stereotypic clinical patterns and frequencies, which increased despite administration of antiepileptic drugs (AEDs). Magnetic resonance imaging (MRI) demonstrated a mass arising from the superior cerebellar peduncle, although video-scalp EEG monitoring revealed no abnormal findings. Positron emission tomography with [18F]fluorodeoxyglucose revealed focal hypermetabolism in the same area identified by MRI. A depth electrode implanted in the mass revealed focal spike-and-wave discharges. The lesion was partly removed; pathologic diagnosis was ganglioglioma. Because of incomplete seizure control and residual tumor visible on MRI, a second operation was performed. After complete excision of the tumor, the patient became seizure free without AEDs. This case confirms the presence of seizure originating from the cerebellum and emphasizes the need for the complete removal of an epileptogenic lesion.

Epileptic seizures are defined as excessive paroxysmal discharges from nerve cells leading to clinical and electrical changes, which usually arise from the cerebral cortex (1). Subcortical structures are thought to have indirect effects only on cerebral epileptic activity through neurochemical modulation (2,3). However, epilepsy of subcortical origin has been demonstrated in animal models (4). More recently, the subcortical origin of epileptic seizures has been well documented in humans with hypothalamic hamartomas (5) and cerebellar dysplastic lesions (6,7).

Depth electrodes placed within hypothalamic hamartomas have revealed ictal activity arising from these structures (5). The epileptogenicity of these hamartomas may come from the aberrant connectivity of neuronal structures or from the intrinsic epileptogenicity of dysplastic neurons (8).

Although most of the cerebellar outputs inhibit epileptic seizure, Jayakar and Seshia (9) reported the case of a child with a cerebellar astrocytoma with paroxysmal segmental rhythmic myoclonus. Marked clinical improvement after tumor resection suggested that the cerebellar lesion was primarily responsible for the movement disorder. Harvey et al. (6) presented the case of a child with hemifacial seizures and cerebellar ganglioglioma and reviewed six previously reported cases of infants with hemifacial spasm and cerebellar masses. In the single case they presented, single-photon-emission computed tomography (SPECT) and a depth electrode revealed that hemifacial seizures can originate in the cerebellar ganglioglioma itself. In addition to ganglioglioma, dysplastic lesions such as gangliomatous hamartomas were suggested to have epileptic seizure activities in the cerebellum (7).

We describe the case of an infant with cerebellar ganglioglioma and episodic paroxymal facial spasms, shown to be of cerebellar origin. Our case provides more evidence confirming the existence of epilepsy of cerebellar origin.

CASE REPORT

  1. Top of page
  2. Abstract
  3. CASE REPORT
  4. DISCUSSION
  5. REFERENCES

A 4-month-old male infant had a history of paroxysmal episodic facial contractions. He was born after a 39-week gestation by normal uneventful vaginal delivery and weighed 3.75 kg. A few hours after birth, paroxysmal facial contractions developed, originating mainly from the left forehead and around the left orbicularis oculi muscles. Paroxysmal hyperventilations with or without abnormal vocalization, nystagmoid eyeball movements, or forceful blinking were occasionally seen. During some episodes, involuntary tremulous movements and brief elevations or swimming-like movements of the limbs were observed, but these were not consistently lateralized. These abnormal paroxysmal movements evolved to stereotypic clinical patterns over a period of weeks, and their frequency increased, lasting for 10 to 60 s. These were repeated ∼100 times per day. They were not provoked by recognizable factors and occurred while patient was awake and asleep. During and after each episode, the baby was alert and attentive and interacted normally. In spite of these recurrent paroxysmal movements, psychomotor developmental milestones were normal, with head control and a social smile. Neurologic examination showed normal posture, tone, and strength. The patient had no hand preponderance, abnormal reflexes, or resting tremors, and there was no evidence of autonomic dysfunction. The physical findings were normal.

A brain magnetic resonance imaging (MRI) at age 4 months showed a mass sized 1.5 × 1.0 × 1.0-cm in the left superior cerebellar peduncle, effacing the fourth ventricle (Fig. 1). No abnormal findings were noted in the cerebrum or brainstem. Repeated interictal electroencephalogram (EEG) recordings showed no abnormalities. Video-EEG monitoring with conventional scalp electrodes revealed no abnormal findings during the interictal, ictal, or postictal state. Positron emission tomography (PET) with [18F]fluorodeoxyglucose (18F-FDG) revealed focal hypermetabolism in the lesion of the left cerebellar peduncle, that is, in the same area identified by brain MRI (Fig. 1).

image

Figure 1. Features of magnetic resonance imaging (MRI) and positron emission tomography with [18F]fluorodeoxyglucose (18F-FDG). Brain MRI at age 4 months shows a 1.5 × 1.0 × 1.0-cm mass in the left superior cerebellar peduncle, effacing the fourth ventricle (left). Notice the marked 18F-FDG uptake of the cerebellar mass, the same region identified by brain MRI (right).

Download figure to PowerPoint

When he was between ages 4 and 8 months, the patient was treated with phenobarbital (PB), valproate (VPA), and clobazam (CLB), but the frequency and duration of the abnormal movements increased, and motor development was arrested. In view of the unsuccessful response to antiepileptic drugs (AEDs) and the findings of brain MRI and PET scans, the origin of the abnormal paroxysmal movements was thought to be in the cerebellar mass lesion. A four-channel depth electrode was inserted into the mass, and two six-channel strips were placed on the right cerebellar hemisphere (Fig. 2). Because of moderate brain swelling after these procedures, subdural strips were not placed on the left cerebellar hemisphere. During continuous extraoperative intracranial monitoring, numerous typical episodes were recorded with stereotypic electrophysiologic changes: 1.5- to 2.0-Hz spike-and-wave discharges following focal 6- to 7-Hz repetitive spike discharges from the electrode located in the mass of the cerebellar peduncle (Fig. 2).

image

Figure 2. Extraoperative electrophysiologic recording (upper) using a four-contact depth electrode in the cerebellar mass and two six-contact strips on the right cerebellar hemisphere, as shown in the skull radiographs (lower). During continuous extraoperative intracranial monitoring, numerous typical episodes were recorded with stereotypic electrophysiologic changes. Spike-and-wave discharges of 1.5- to 2.0-Hz after focal 6- to 7-Hz repetitive spike discharges from the one electrode located in the mass of the cerebellar peduncle are shown.

Download figure to PowerPoint

When the patient was 8 months old, the mass was partially resected to avoid brainstem injury. A pathologic diagnosis of low-grade ganglioglioma was made. Repeated brain MRIs after surgery revealed a small residual tumor. After the first operation, the frequency of seizures decreased but was not completely controlled. The residual mass was completely resected at age 11 months, and histopathologic examination revealed a ganglioglioma without invasion of the adjacent tissue. Immediately after surgery, the patient showed ataxia and weakness of the left extremities. Thirty months after the second operation, the patient had mild ataxia. No seizures occurred even after the discontinuation of AEDs.

DISCUSSION

  1. Top of page
  2. Abstract
  3. CASE REPORT
  4. DISCUSSION
  5. REFERENCES

Hemifacial spasms in infants with a cerebellar mass were first described by Langston and Tharp (10). Since then a small number of similar cases of a cerebellar or fourth ventricular mass have been reported, with a manifestation of hemifacial spasm or hemisomatic spasms (9,11–12).

This case involves a patient with identical hemifacial spasms and cerebellar ganglioglioma. The absence of lost consciousness, the presence of nystagmus, and the absence of EEG abnormalities during the ictal period led us to speculate that the seizures were of cerebellar origin.

To investigate seizure origin, we performed PET and invasive electrophysiologic monitoring. PET demonstrated the presence of focal hypermetabolism, which was in concordance with the MRI findings. The hypermetabolic finding on PET was somewhat unexpected because the epileptogenic zone generally exhibits hypometabolism on interictal PET. This could be explained by persistent seizure activity during the study, or by the metabolism of the tumor. Meyer et al. (13) reported high 18F-FDG uptake in mixed neuronal–glial tumors, such as gangliogliomas. Di Chiro et al. (14) also stated that high-energy metabolism in gangliogliomas may be expected because they consist at least partially of neuronal cells. Invasive electrophysiologic monitoring with a depth electrode also confirmed the cerebellum as the area of seizure origin. Depth electrodes placed within gangliogliomas have previously revealed ictal activity arising from these structures, which correlates with hemifacial spasms in our patient as well as in other reported patients (6).

The main surgical consideration in this case was the extent of tumor resection, as there was some reluctance to undertake total resection because of possible brainstem injury. Unfortunately, partial resection did not completely resolve the seizures. Sequential complete resection achieved optimal seizure control and the prevention of tumor recurrence without significant neurologic deficit. We are of the opinion that complete resection of a tumor provides the most important condition for seizure relief, in addition to preventing tumor recurrence.

We believe that the case described provides sufficient evidence to suggest that epileptic seizures originated from the subcortical structures. Clinical features, functional imaging, and invasive electrophysiologic study indicated that the seizures originated from the cerebellar lesion. Complete seizure control after resection of this lesion confirmed that the seizures were primarily cerebellar in origin. This case also emphasizes the importance of complete lesionectomy for optimal seizure control in cases involving intractable epilepsy associated with brain tumor.

Acknowledgment: This study was supported by a grant from the Korea Health 21 R&D Project, Ministry of Health & Welfare, Republic of Korea (HMP-99-N-02-0003).

REFERENCES

  1. Top of page
  2. Abstract
  3. CASE REPORT
  4. DISCUSSION
  5. REFERENCES