Ictal FDGPET and SPECT in hemifacial seizures due to cerebellar epilepsy—Case report
Correspondence Address: Source of Support: None, Conflict of Interest: None DOI: 10.4103/0028-3886.253622
Source of Support: None, Conflict of Interest: None
Keywords: Cerebellar, epilepsy, FDG-PET, hemifacial seizures, SPECT
Epilepsy mainly arises from the cerebral cortex. Subcortical lesions are generally considered coincidental. A subcortical origin, though debatable, has been postulated with reports of seizure onset from a hypothalamic hamartoma  and cerebellar dysplastic lesions., Epileptogenicity of subcortical structures is postulated from either aberrant connectivity of neuronal structures or from the intrinsic epileptogenicity of dysplastic neurons. Interictal and ictal electroencephalography (EEG) data are mostly non-contributory in these cases, posing difficulty in the pre-surgical workup. We report the findngs on two operated cases of cerebellar refractory epilepsy and highlight the role of ictal (18) F-fluorodeoxyglucose–positron emission tomography (FDG-PET) and ictal single photon emission computed tomography (SPECT) in establishing the epileptogenicity of the cerebellar lesion.
A 4-month old male infant had seizures since 1 month of age with a frequency of 6 to 8 seizures per day, which increased to 15 to 20 per hour over the next 1 month. The seizure began with left eye blinking, followed by left face jerking and head and eye deviation to the left, with dystonic posturing of the left upper limb and lower limb for 1 to 3 minutes. He had hypersalivation with tachypnea during these episodes. The child was unresponsive for a few minutes and then became playful. Transient post ictal weakness of the left upper and lower limbs was noted. He failed multiple antiepileptic drugs (AEDs). He attained a social smile. His birth history was unremarkable. The inter-ictal EEG showed right hemispherical 2-3 Hz slow wave activity. Ictal EEG showed a right hemispherical 2-3 Hz slow wave activity. Ictal EEG showed right hemispherical 2-3 Hz slow wave activity during the entire event. Magnetic resonance imaging (MRI) of the brain revealed a non-enhancing left cerebellar mass involving the uvula, pyramid, medial aspect of middle and superior cerebellar peduncle, and inferior colliculus, extending up to the midline [Figure 1]a and [Figure 1]b. Ictal SPECT revealed focal lesional hyperperfusion while ictal FDG-PET study showed focal hypermetabolism [Figure 1]c and [Figure 1]d. The child underwent bilateral sub-occipital craniotomy with near-total excision of the left cerebellar lesion. Rhythmic spike discharges were noted on electrocorticography (ECoG) during depth electrode monitoring. The histopathology was a low-grade ganglioglioma. The child was seizure-free for 6 months, and at a 6-year follow-up, getting occasional brief left hemifacial jerks. He was going to school, able to communicate well, and was on three anti-epileptic drugs (AEDs).
A 2-year and 2-month-old child presented with seizures since 10 days of birth. Seizures start with right eye blinking and right face jerking lasting for a few seconds. Initially, the frequency was 5 to 10 per day, which increased to 5 to 6 per hour. His birth and developmental history was normal. Interictal EEG was suggestive of occasional left frontal and parasagittal sharp wave discharges. The ictal EEG was suggestive of left frontal and parasagittal ictal onset. MRI revealed a right cerebellar, right middle cerebellar peduncle, juxta- and intra-fourth ventricular space occupying lesion [Figure 2]a and [Figure 2]b. Ictal FDG-PET CT was suggestive of hypermetabolic focus in the area of the lesion and ictal SPECT revealed hyperperfusion [Figure 2]c, [Figure 2]d and [Figure 2]e. Epileptiform discharges were recorded during intra-operative ECoG with depth recordings from the lesion. The child underwent surgery, and resection of the lesion was done. Post surgery, he was completely seizure-free at a 4-year follow-up with normal milestone development and was going to school. Histopathology was suggestive of a low-grade glioma.
The two cases highlight that refractory epilepsy may occur due to a cerebellar lesion. They highlight the importance of PET and SPECT imaging in confirming the epileptogenicity of the lesion. In 1871, John Hughlings Jackson first described a 5-year-old boy with a tuberculous abscess of the middle lobe of cerebellum, presenting with tonic seizures, introducing for the first time the concept of “cerebellar seizures” that was distinct from “cerebral epilepsy”. Harvey et al., in their seminal paper proposed a unique clinicopathologic syndrome of hemifacial seizures and cerebellar ganglioglioma occurring in infancy. During the late half of the 20th century and early 21st century, many authors published clinical cases of seizures in the setting of cerebellar tumors., Recently, Foit et al., had reviewed 29 case reports of a total of 31 patients with lesional cerebellar epilepsy.
While the role of hypothalamic hamartoma in epilepsy is well established, the role of cerebellar glioneuronal tumors (GNTs) is considered elusive. This is due to various reasons. First, the cerebellum is a subcortical structure with low epileptogenic potential. It is supposed to exert a modulatory effect on epileptic seizures, rather than being the primary seizure generator. Second, the routine pre-surgical evaluation including the semiology characterization and EEG evaluation has a poor or almost negligible contribution in defining the role of cerebellum in causing epilepsy. Scalp EEG, assessing both interictal and ictal EEG, is mostly normal or non-localizing in these cases. Neither is the semiology a characteristic one. Different types of seizure semiology can be seen. Hemifacial seizures are the most common manifestations and can be confused with movement disorders. Myoclonic seizures and focal or generalized seizures are other uncommon seizure types. Myoclonic and generalized seizures do not give a clue to lateralization, and even in focal seizures, the semiology is paradoxically ipsilateral to the side of lesion (e.g., a left-sided cerebellar lesion can cause left focal seizures). All these factors pose a difficulty in proving the role of cerebellar lesions in causing epilepsy during the non-invasive pre-surgical evaluation.
Nuclear imaging studies in the recent times may help to circumvent this problem. Positive FDG-PET and ictal SPECT have been shown to be highly suggestive of the propagating epileptic activity, providing complementary pre-surgical information that matched video-EEG results and clinical data in cerebral seizures., This utility of nuclear imaging is even more useful in cases of cerebellar epilepsy where video-EEG results and clinical data may fail to prove its epileptogenicity. In both our cases, ictal FDG-PET demonstrated well-localized hypermetabolism concordant with the MRI lesion. This was demonstrated earlier on, in a few studies. Lascano et al., reported a 14-month-old girl presenting with hemifacial spasm and employed electric source imaging, which was in concordance with the nuclear imaging findings, and subsequently validated by the intra-operative intralesional recordings to prove the epileptogenicity of cerebellar lesion. They suggested that hemispheric cerebellar lesions in cases like this are likely to cause epilepsy via an effect on the facial nuclei through ipsilateral and contralateral aberrant connections. Similar findings were reported in another study of a 4-month-old infant with paroxysmal facial contractions; the positron emission tomography with (18) F-FDG revealed focal hypermetabolism in the same area identified by MRI, subsequently proven by invasive recordings. On the same lines, ictal SPECT showed focal hyperperfusion in the area of the lesion in both our patients, similar to the few cases reported previously. Hanai et al., had reported a case of refractory hemifacial seizures in which MRI revealed a large tumor in the left middle cerebellar peduncle, and ictal SPECT and ictal (18) F-FDG PET revealed hyperperfusion and hyper-glucose metabolism at the tumor site. Total removal of the tumor resulted in complete disappearance of hemifacial seizures and improved psychomotor development, indicating that the cerebellar tumor caused hemifacial seizures. In all these studies, the semiology and EEG data were not useful in localizing the epileptogenic zone, and epileptogenicity was finally proven by invasive recordings. Considering all this information and our experience, we hereby postulate that ictal FDG-PET and ictal SPECT studies will confirm the epileptogenic nature of the cerebellar lesions and obviate the need for an invasive evaluation.
Glial-neuronal tumors (GNTs) and hamartomas are responsible for the most common lesional cerebellar epilepsies reported. Other common lesions include low-grade gliomas. Evidence is growing that cerebellar tumors have dysplastic neurons which are inherently epileptogenic. Whether intrinsic epileptogenicity  or peritumoral focal cortical dysplasias (FCD) are the focus of ictal onset is uncertain. Some even suggest the term 'dysplastic cerebellar epilepsy' as it is felt that GNTs and FCD form a continuum. The outcome of surgical management of cerebellar epilepsy is favorable, with about 78% patients achieving an Engel I outcome at a 3-month follow-up. The best timing of surgery is not known.,
Another important issue is that epilepsy is not seen in all cerebellar lesions or tumors. One of the reasons cited is the location within the cerebellum. The reported cases of “cerebellar epilepsy” had lesions close to cerebellar peduncles, similar to the two cases in the present report. Proximity to cerebellar peduncles, which are the main efferent fiber tracts, may be important for plausible epileptogenic connections with other areas of the brain. It would be interesting to study the role of nuclear imaging in predicting which cerebellar lesions will be epileptogenic.
Lesions of subcortical structures like the cerebellum can cause epilepsy especially when they are in close proximity to the peduncles. Ictal FDG-PET and ictal SPECT studies are important non-invasive methods in the pre-surgical evaluation to confirm the epileptogenicity of the cerebellar lesion.
Declaration of patient consent
The authors certify that they have obtained all appropriate patient consent forms. In the form the patient(s) has/have given his/her/their consent for his/her/their images and other clinical information to be reported in the journal. The patients understand that their names and initials will not be published and due efforts will be made to conceal their identity, but anonymity cannot be guaranteed.
Financial support and sponsorship
Conflicts of interest
There are no conflicts of interest.
[Figure 1], [Figure 2]