Epilepsia Partialis Continua

Updated: Jun 13, 2023
  • Author: Elanagan Nagarajan, MD, MS; Chief Editor: Selim R Benbadis, MD  more...
  • Print



Epilepsia partialis continua (EPC) was a name first introduced by Kojewnikoff on January 21, 1894 when he presented his description of a unique type of prolonged focal seizure in four patients at a meeting of the Moscow Neurological and Psychiatric Society. [1] Since then, this name has been ascribed to various nosological entities, with the condition being described as a variation of jacksonian epilepsy, jacksonian status epilepticus, myoclonus epilepsy, an extrapyramidal syndrome, or continuous, localized myoclonia.

EPC can be considered the status epilepticus (SE) equivalent of focal onset seizure with retention of awareness. It can manifest as focal motor or sensorimotor clonic seizures without jacksonian march. Seizures remain localized to the part of the body in which they originate, and motor activity is often persistent, even during sleep, lasting for at least 60 minutes and  up to hours, days, weeks, or even longer. Consciousness is usually preserved, since the thalamus is spared and seizures remain unilateral in most cases. Postictal weakness (also known as Todd’s paralysis) is frequently evident. [2, 3]

In the vast majority of cases, the seizures are of cortical origin; however, subcortical mechanisms also have been proposed. [4] EPC can be an expression of a stable neurologic lesion or a progressive disease, such as Rasmussen syndrome. [5] Antiepileptic drugs, with a few notable exceptions, do not seem to significantly alter the course of this condition. [6] While the localized nature of EPC is one of its most striking features, occasional spread is seen in otherwise typical cases, and using spread as an exclusion criterion is not warranted.

Historical perspective and definition

Kojewnikoff (1895), [7] Bruns (1895), [8] Orlovskij (1896), [9] and Choroschko (1907) [10] all reported a similar condition characterized by intermittent muscular twitching of a body part while consciousness is unaffected. Kojewnikoff described 4 cases in which the patient experienced seizures, consisting of constant jerking movements of high frequency that were resistant to treatment and involving only one part of the body. [1] He recognized that this type of seizure could have many causes, such as a tumor, an abscess, syphilis, edema, embolism, or localized encephalitis. He postulated, without postmortem studies, that the seizures resulted from focal cortical encephalitis and localized the inflammation of the brain in the motor pathway. [1, 7, 11] This original description of EPC has been extensively discussed, [12, 13] and its incidence in Russian spring-summer encephalitis has been recently reviewed. [14]

From 1904-1907, Speilmeyer and Mills reported, [15, 16] under the name of jacksonian epilepsy, cases in which EPC may have been present. Since then, a number of clinical and pathological studies have further defined the syndrome, its location, and its etiology. [17]

In 1966, Juul-Jensen and Denny-Brown defined EPC as “clonic muscular twitching repeated at fairly regular short intervals in one part of the body for a period of days or weeks.” [4] They also differentiated it from myoclonus by characterizing the latter as “more rapid, lasting less than half a second, involving a variable number of muscles... at irregular intervals,” but recognized the existence of transition forms, which are difficult to classify.

In 1977, Thomas et al defined EPC as “a partial somatomotor SE... for a minimum of one hour and recurring at intervals of no more than 10 seconds.” [17] They carefully described the features of EPC in 32 patients. EPC had a sudden onset in 17 of 25 patients; its durations varied from 4 hours to 18 years. Movement frequency was 0.5-10 Hz, and the duration varied from 25-600 milliseconds, with twitches often recurring every 2-3 seconds. Spike or sharp wave on scalp EEG preceded the movement in 9 of 12 patients, and all 5 corticographies revealed a cortical origin. Seventeen of 25 patients had a poor response or no response to anticonvulsants, and 6 of the 7 responders had acute brain lesions so that the therapeutic success could have been coincidental. Interestingly, cortical excision failed to stop the seizures in 4 of 5 patients.

In 1983, Gastaut defined EPC as a subtype of somatomotor focal with retained awareness SE, which was itself a form of elementary focal SE; this definition made the point that there are as many types of SE as there are types of seizures. [18] Several subsequent reviews have used this definition. [6]

In 1985, Obeso et al defined epilepsia partialis continua as spontaneous regular or irregular clonic muscle twitching of cerebral cortical origin, confined to one part of the body and continuing for a period of hours, days, or even weeks. [19] The addition of “irregular” to the description recognizes the fact that the interval between motor twitches can vary but, in the author’s view, makes a clear distinction from myoclonus impossible.

Cockerell et al (1996) [20] and Shorvon (1994) [21] defined EPC as “a syndrome of continuous focal jerking of a body part occurring over hours, days, or even years” and restricted the definition to jerks of cortical origin; they proposed the name “myoclonica continua” for jerks of subcortical origin. While this distinction is attractive in principle, in practice localizing the exact origin of the seizures is often difficult, and clinical activity indistinguishable from EPC has been associated with subcortical pathology (eg, multiple sclerosis).

Wilson and Winkelman, (1924) [11] Hess and Sethi, (1990) [22] and Biraben and Chauvel (1998) [23] restricted EPC to local and elementary motor signs, regardless of whether it was associated with spreading seizures at times.

In current practice, EPC is defined as a form of focal SE with simple motor manifestations that are maintained for over 1 hour, with clonic activity restricted to one body part and recurring at fairly regular intervals. The following modifying factors apply:

  • Motor activity often is modified by sensory stimuli

  • Frequency is usually 0.1-6 Hz

  • Epilepsia partialis continua can continue for long periods (sometimes years) without spreading, although spread can occur at times

  • Epilepsia partialis continua often is associated with postictal or interictal weakness

Clonic activity in EPC can involve any muscle group and is most common in the upper extremities. Although it is typically invariant and remains localized to a single muscle group in most patients, it may be accompanied by jacksonian spread of the seizure, which may even lead to a focal seizure with loss of awareness or secondarily generalized seizure. This syndrome also may be accompanied by other neurological and psychopathological symptoms.

Bancaud classification

Bancaud et al (1982) [3] classified EPC into 2 groups. Both entities start with similar seizures, but type 2 proves to be intractable and progressive. See the Table below.

Table. Bancaud Classification (Open Table in a new window)

Type 1 ( Classic)

Type 2 (Rasmussen)

Rolandic fixed lesion

Normal developmental and history until seizure onset

Neurological deficit

Preceding focal motor

Preceding focal motor seizures

Following myoclonic jerks

Following myoclonic jerks

Abnormal EEG background with focal and diffuse paroxysmal abnormalities

Focal abnormalities on EEG

Progressive course

Nonprogressive course

Intractable epilepsy

Surgery usually effective


Bancaud divided the clinical course of type 2 into 3 stages. The first stage features only focal motor seizures with retained awareness or focal seizures with loss of awareness, but EPC may occur. In the second stage, EPC is seen in the setting of progressive neurologic deficit and mental deterioration. The third stage is characterized by arrest of deterioration and decrease or disappearance of seizures.

Go to Epilepsy and Seizures, Status Epilepticus, and Partial Epilepsies for an overview of these topics.

Patient education

For patient information, see the Brain and Nervous System Center, as well as Epilepsy.



Since the first description by Kojewnikoff, a number of isolated case reports and small case series have elucidated this syndrome further. However, few epidemiologic studies have been performed. Cockerell et al (1996) estimated the prevalence of EPC at less than 1 case per million population, based on 36 cases reported in the United Kingdom over a 1-year period, 10 of the cases being new. [20]

The incidence of EPC is slightly higher in males than in females. [4, 17, 20] In a retrospective review by Sinha and Satishchandra (2007) of 76 patients with EPC at a tertiary care center in South India over the course of 14 years, the investigators found a male-to-female ratio of 46:30, a mean age of 30.2 ±23.4 years, and a median age of 26 years. [24]



Historical studies

In 1895, Kojewnikow postulated that myoclonus and seizures arose from the cerebral cortex in association with localized encephalitis. His brilliant insight contrasted with the then current view that somewhat clinical entities, such as Friedreich paramyoclonus multiplex and familial progressive myoclonic epilepsy of Unverricht, resulted from hyperexcitability of the anterior horn cells. [25] and it has been proven right in the overwhelming majority of cases of EPC.

In 1922, Souques considered that dysfunction of the motor cortex was the cause of focal motor seizures, but that localized myoclonias must be of subcortical origin. [26] The first EEG study of myoclonus with epilepsy was performed by Grinker et al. [27] They demonstrated a temporal correlation between frontal polyspikes and waves and myoclonic jerks at the periphery.

In 1947, Dawson showed that the motor jerks were preceded by fast (30 Hz) discharges in the contralateral motor cortex. [28] The EEG discharge had an exclusive waveform, beginning with a positive wave immediately followed by negative shift on which a train of spikes was superimposed. When neither the negative shift nor the spike train occurred, no myoclonic jerking was observed.

Dawson also recorded scalp somatosensory evoked potentials (SSEPs) for the first time in response to the electrical stimulation of the nerve trunk. Their maximal amplitude was located near the midline for the lower limb, and 6-8 cm more lateral to the upper limb. He concluded that these evoked potentials represented exaggerated responses of the underlying cortex to peripheral stimuli, and that hyperexcitability of motor cortex, rather than increase of afferent volley to the primary somatosensory cortex via subcortical mechanisms, would account for the precentral site of projection.

In 1954, Kugelberg and Wieden reported a case of EPC and remarked on the constant relationship with a 27-34 millisecond latency, between myoclonic jerking of the right foot and the focal EEG spikes in the central areas, maximal on the left side near the midline. [29] In this patient, the motor cortex was excised with complete remission of the seizures.

Cockerell et al. provided compelling evidence based on electrographic studies to suggest a relationship between cortical reflex myoclonus and EPC. [20]  In addition to electrographic evidence, they observed activation of EPC by external and proprioceptive stimulation.

Guerrini et al. highlighted a possible role of cortical reflex myoclonus transitioning into cortical seizure fragments then into continuous seizure activity. [30] A suggestion was raised to further expand the relationship between cortical myoclonus to EPC to include cortical tremors; however, the frequency of 7–15 Hz of movements in cortical tremors is different from EPC. This raises the question as to whether these entities exist on a spectrum with cortical tremors representing the milder end and EPC on the more severe end of the spectrum. Such observations further highlight the variability of presentation and underlying mechanism of EPC.


The unequivocal cortical origin of EPC appears to be substantiated in humans by clinical, electrophysiologic, and neurosurgical evidence. In addition, an epileptogenic lesion of the central neocortex can mimic this phenomenon in monkeys. [31]

The neocortex is endowed with powerful lateral inhibition, which probably is designed to keep responses precisely localized. This may be the reason that EPC can go on for a long period while remaining precisely localized to a small group of muscles and a small cortical domain. Since surround inhibition may limit seizure spread more effectively in the motor neocortex than in any other area because of the tight afferent-efferent relationships, which support the activation of long-loop reflexes, EPC may be a unique expression of cortical organization. [32, 31] However, note that seizures sometimes can be driven from distant, or even subcortical, sites. [33]

By contrast, the limbic system, which is involved heavily in processes such as memory and emotion, is designed to spread excitation widely and to modulate excitability in many brain regions. Perhaps, as a result of this organization, seizures of limbic origin often spread widely and rapidly.


The EPC focus (1) is hypermetabolic, as shown by 2-deoxyglucose positron emission tomography (PET) scanning; (2) is hyperactive, as seen as by magnetoencephalogram (MEG) studies; and (3) shows increased blood flow in single-photon emission computed tomography (SPECT) scan studies. [23]

Synaptic mechanisms

Synaptic mechanisms have been studied only in self-sustaining SE, not in EPC. The author now suspects that self-sustaining SE is initiated by failure of GABAergic inhibition but is maintained by widespread potentiation of excitatory (especially N -methyl-D-aspartate [NMDA]) synapses; therefore, established self-sustaining SE becomes resistant to all agents except NMDA antagonists. [34] The authors speculate on a similar mechanism in EPC, in which the focus would be characterized by long-term potentiation of glutamate receptors and desensitization of GABA receptors, while GABAergic inhibition would be preserved in the surround.


Clinical Presentation

Clonic activity can involve any muscle group and is most common in the upper extremities. Although it is typically invariant and remains localized to a single muscle group in most patients, it may be accompanied by jacksonian spread of the seizure, which may even lead to a focal seizure with loss of awareness or secondarily generalized seizure. About 60% of patients have other types of seizures besides EPC. [20] The presence of concomitant seizure types with EPC has been illustrated in multiple studies  [17, 20, 24, 35, 36] that reported focal seizures with and without loss of awareness as well as secondarily generalized seizures manifesting together with EPC. A study by Kravljanac et al, [37] looked at 51 pediatric patients all of whom experienced one or more types of seizures besides EPC. This syndrome also may be accompanied by other neurological and psychopathological symptoms.

The topographic distribution of seizures from one study of EPC is as follows:

  • Motor deficit (56%) - Tetraparesis (4%), hemiparesis (23.3%), monoparesis (13.3%), oculomotor paresis (0.7%), facial paresis (10%), hypoglossal paresis (1.9%), soft palate paresis (0-3%)

  • Sensory deficit (21%) - Hemihypoesthesia and hemianesthesia (9.7%), monohypoesthesia and monoanesthesia (4.4%), deep sensation (3.6%), stereognosis (3.3%)

  • Anisocoria (1%)

  • Conjugate deviation of eyes (1%)

  • Homonymous hemianopsia (5%)

  • Aphasia (14%) - Motor (8.3%), sensory (2.5%), amnestic (2.6%)

  • Apraxia (1%)

  • Alexia (2%)

Neurological symptoms distribution in a study of epilepsia partialis continua (N=246) is as follows:

  • Head (n=24) - Mouth (9), face (11), periorbital muscle (2), soft palate (2), hypersalivation (1)

  • Upper extremities (n=60) - Thumb (4), finger (6), hand (24), arm (26)

  • Trunk (n=8) - Shoulder (4), breast (1), abdomen (3)

  • Lower extremities (n=22) - Big toe (2), leg (10), foot (10)

Neuropsychological symptoms distribution in a study of epilepsia partialis continua is as follows:

  • Disturbance of consciousness (36%) - Somnolence (19%), stupor (3%), coma (14%)

  • State of confusion (15%)

  • Disorientation (10%)

  • Labile affect (5%)

  • Stupor (5%)

  • Dementia (15%)

  • Others (14%)

Clinical neurophysiology

Neurologic evaluation is essential in view of the variety of etiologies and associated illnesses in patients with EPC.

Reported EEG abnormalities in EPC include spikes, sharp waves, or slow-wave activity with periodic lateralized epileptiform discharges. [35] The 2011 European EPC Survey of 65 cases found epileptiform activity in 42 cases, slow waves in 11 cases, and local flattening in one case. [36]  Eleven patients were reported whose EEG was unrevealing, however, a normal EEG does not exclude the diagnosis of EPC as it can be normal in 20% of patients with EPC. [17, 20]  Failure of EEG to demonstrate epileptiform discharges or focal abnormalities may imply that the area of cortical involvement is too small to record or of a subcortical origin. In Rasmussen encephalitis, the EEG often shows significant lateralized, slow-wave activity comprised of polymorphic delta waves in involved hemisphere, and it may give evidence of other seizure types suggestive of widespread, but lateralized, disease. Frequently, lateralized, asymmetrical background slowing is noted.

Similar findings also can be seen with other etiologies, such as glial tumors. In the case of focal, nonprogressive pathologies, such as chronic stroke, the background is rarely as abnormal or asymmetrical. This point may help in differentiating a focal, nonprogressive pathology from structural diseases with a coexistent metabolic encephalopathy, in which the background is often diffusely abnormal.

Evoked-potential techniques, especially somatosensory evoked potentials (SSEPs), have been used to examine the physiologic mechanisms and anatomical locations of EPC. [6] Giant SSEPs are seen often and point to cortical hyperexcitability, which may be an essential mechanism of EPC. SSEPs have also been useful in providing evidence of EPC being related to cortical origin reflex myoclonus. [20]



Clinical and pathologic studies have identified various presumed causes of EPC. This long list reflects the diverse nature of the pathologic processes that cause focal neocortical seizures. Nothing about these processes is specific, and the unique clinical features of EPC are likely to reflect its anatomic location rather than its etiology.

In recent years, autoimmune processes associated with Rasmussen encephalitis have received considerable attention. [38, 39]  A rare and chronic progressive inflammatory disease, Rasmussen encephalitis usually affects one cortical hemisphere in children, although cases of later onset have been reported. [40]

Idiopathic Epilepsy

  • Benign Childhood Epilepsy with Centro-temporal Spikes (Benign Rolandic Epilepsy)

Cerebral neoplasia–related causes are as follows:

  • Metastasis

  • Astrocytoma

  • Oligodendroglioma

  • Carcinomatosis cerebri

  • Hemangioma

  • Lymphoma

  • Meningioma

Cortical dysplasia–related causes are as follows:

  • Conical dysplasia

  • Hemimegalencephaly

  • Tuberous sclerosis

  • Linear sebaceous nevus syndrome

  • Sturge-Weber syndrome

Traumatic lesion–related causes are as follows:

  • Acute head trauma

  • Posttraumatic cyst

  • Subdural hematoma

  • Intracerebral hematoma

  • Epidural hematoma

Drug-induced causes are as follows:

  • Penicillin

  • Azlocillin

  • Cefotaxime

  • Metrizamide

Metabolic causes are as follows:

  • Diabetic ketoacidosis

  • Nonketotic hyperglycemia

  • Hepatic encephalopathy

  • Uremic encephalopathy

  • Hyponatremia

Infectious or parasitic causes are as follows:

  • Abscess

  • Tuberculoma

  • Gumma

  • Russian spring-summer encephalitis

  • Subacute measles encephalitis

  • Human immunodeficiency virus

  • Progressive multifocal leukoencephalopathy

  • Creutzfeldt-Jakob disease

  • Viral encephalitis or meningoencephalitis

  • Cryptococcal meningitis

  • Anti-Hu–associated paraneoplasia

  • Cysticercosis

  • Granulomatous diseases

  • Pertussis infection

  • COVID-19 pneumonia [41]

Vascular lesion–related causes are as follows:

  • Arteriosclerotic cerebrovascular disease

  • Embolic or postthrombotic ischemic infarction

  • Cortical venous thrombosis

  • Cerebral hemorrhage

  • Systemic lupus erythematosus

  • Sjögren syndrome

  • Arteriovenous malformation

  • Carotid hypoplasia

Autoimmune causes are as follows:

  • Rasmussen chronic encephalitis

  • Multiple sclerosis

  • Anti-GluR3 or anti-NMDA-GluR-Epsilon 2 antibodies

Genetic causes are as follows:

  • Alpers disease

  • Kufs disease

  • Leigh syndrome and cytochrome C oxidase deficiency

  • Nicotinamide adenine dinucleotide (NADH) coenzyme Q reductase deficiency

  • Mitochondrial cytopathies including mitochondrial encephalopathy with lactic acidosis and stroke (MELAS)



Positron emission tomography (PET) scanning and single-photon-emission computed tomography (SPECT) scanning are emerging as useful research tools in the evaluation of the metabolic effects of EPC, especially when computed tomography (CT) and magnetic resonance imaging (MRI) findings are normal or may show a lesion.

PET scan studies are also likely to enhance understanding of the biochemical and metabolic features associated with the abnormal physiology, and, consequently, they may provide significant information not only for the diagnosis of EPC but also for its treatment. [42] Burneo et al. reported that ictal SPECT has also been useful in the presurgical evaluation of Rasmussen encephalitis. [43]

MRI can point to the structural lesion of the cortex and/or white matter, and it can follow the progression of atrophy in Bancaud type 2 EPC. [23] With easier access to MRIs and availabilities of higher resolution imaging modalities, radiographically subtle etiologies of EPC, such as focal cortical dysplasia, are more frequently reported, as highlighted by a European EPC survey which reports cortical dysplasia in 13.8% of EPC cases. [36]

SPECT-MRI fusion has been reported by Matthews to have been successfully used to identify epileptic focus in a patient with EPC. [44]



The long-term prognosis of EPC depends on its underlying cause. The early onset of EPC in a child is often, but not always, a manifestation of neurodegenerative disease or Rasmussen encephalitis, and these conditions often are associated with progressive neurologic decline. Numerous studies have suggested that EPC is not in the infectious process but has autoimmune and inflammatory components. Cytokines, such as interleukin 1B and tumor necrosis factor-α, may play a role in EPC as they alter neuronal excitability and contribute to neuronal cell loss, astrogliosis, and blood-brain barrier damage. [45, 46, 47]

Kravljanac et al (2013) conducted a cohort study of 51 children with EPC and recorded neurological consequences in 64.7% and fatal outcomes in 15.7% of the patients. [37] The study found that the strongest predictors for both neurological consequences and lethal outcome were EEG characteristics, the application of intensive therapy, and EPC onset in the late phase of disease. A better prognosis was reported in patients with EPC type I than with Rasmussen encephalitis (EPC type II). [35]

In patients with adult-onset EPC, the underlying cause can be fixed (eg, cortical dysplasia, stable arteriovenous malformation), self-limiting (eg, trauma, stroke), or progressive (eg, tumor, carcinomatous meningitis); the prognosis depends on the underlying pathology.

Kim et al (2012) reported a patient who developed alien hand syndrome after experiencing EPC. [48] Brain MRI and fluorodeoxyglucose (FDG) PET studies found extensive hypometabolism over lesions in epileptogenic zones. Studies by Weitemeyer et al (2005) and Van Paesschen et al (2007) showed high concordance rates between hypometabolism on interictal PET and frontal lobe epilepsy. [49, 50] Thinning of the corpus callosum was noted on MRI. The investigators suggest that EPC may have been the cause of the structural damage to the patient’s corpus callosum resulting in the neurological deficits. Hence, Kim et al, (2012) recommend that physicians consider aggressive and prompt management for certain etiologies of EPC. [48]


Pharmacologic Treatment

Treatment should focus on the underlying condition. Antiepileptic drugs (AEDs) are usually tried to prevent the spread of seizures, but with a few notable exceptions, they are often unsuccessful in altering the course of EPC. Polytherapy is often complicated by adverse effects and drug interactions. [20] Immunosuppression may have a place in the treatment of Rasmussen encephalitis. Hemispherectomy has been shown to prevent the progression of Rasmussen encephalitis in selected cases, at the cost of sometimes aggravating hemiparesis and other deficits. [51, 52]

Anti-epileptic drugs

Benzodiazepines may provide temporary relief. [36]  Phenytoin or phenobarbital may be more effective than carbamazepine or valproate. [23] Echenne (1997) and Bhidayasiri et al (2012) reported success with felbamate in patients who responded poorly to other anti-epileptic drugs (AEDs). [53, 54]  Topiramate may be successful in EPC with dysontogenetic etiologies. [55]  Nimodipine has been administered successfully in 2 cases of EPC following an acute cerebral event. [56]

Control of EPC may be achieved by a combination of intravenously administered AEDs, including diazepam, pentothal sodium, and valproate. [35] A case report by Eggers et al (2009) illustrates a patient who sustained immediate EPC relief after receiving an intravenous bolus of levetiracetam, even though the patient presented with EPC while taking levetiracetam orally. [57] This study supports the view that supratherapeutic serum concentrations of AEDs may help in some patients. [58]


Oral corticosteroid therapy and immunosuppression may be of some benefit in rare cases. [59] Recent evidence supports a correlation between Rasmussen encephalitis and autoimmune processes. [60, 61, 62] Some Rasmussen patients have antibodies to the glutamate receptor subunit, GluR3. [60] or to NMDA receptors. [63]

Natalizumab, rituximab, tacrolimus and intravenous immunoglobulin (IVIG) have been reported to be effective in Rasmussen encephalitis. [64, 65, 66, 67] Barontini et al (1994) reported the use of gamma globulins, but few lasting benefits have been reported to date. [68] Plasma exchange has been reported to have improved EPC significantly in a few Rasmussen patients. [69]

Other pharmacotherapy

Because cytomegalovirus has been implicated in the pathogenesis of Rasmussen syndrome, McLachlan et al (1996) used ganciclovir and reported that EPC was controlled in one patient with this syndrome. [70] Intraventricular interferon alfa was reported to have completely controlled EPC in another patient with Rasmussen encephalitis. [71]

Adverse effects of drugs should also be considered in the differential diagnosis of EPC to facilitate rapid diagnosis and drug discontinuation. Treatment with medications such as penicillin, azlocillin, and cefpodoxime have been reported to cause partial motor seizures. [72, 73] A case study performed by Aberastury et al (2010) reports a patient experiencing EPC with positive white matter lesions in cortical and subcortical regions on MRI after taking levamisole 1 g/wk for 2 weeks. [74] This patient had no prior family history of metabolic or neurologic diseases, nor did she have any medical conditions besides oral aphthous ulcers for which she was prescribed levamisole.


Other Treatment Techniques

Neurosurgical approaches, such as multiple subpial transections, [75] may be used as a last resort. Hemispherectomy should be considered in cases of Rasmussen encephalitis.

Repetitive transcranial magnetic stimulation has been reported to have lasting success in anecdotal cases and may deserve further evaluation. [76, 77, 78] A study was conducted by Rotenberg et al (2009) that looked at 7 patients with EPC of mixed etiologies and after treating the patients with repetitive transcranial magnetic stimulation over the seizure in either high frequency (20-100 Hz) bursts or prolonged low frequency (1 Hz) trains found that 3 of 7 patients experienced brief (20-30 min) pauses in seizures and 2 of 7 experienced a lasting (≥1 d) pause in their seizures. [79]

A case of successful treatment of facial myoclonus with botulinum toxin has been reported. [80]