Neuroimaging techniques used to evaluate patients with FSE-EPC are evolving rapidly. Currently, high-resolution MRI with fluid-attenuated inversion-recovery (FLAIR) sequences is the best technique to visualize some of the microdysplastic lesions that may predispose to FSE when standard MRI fails to show discrete abnormalities. This is also true for cortically restricted abnormalities seen in mitochondrial diseases. In cases in which the suspected underlying pathology has caused breakdown in the blood–brain barrier (e.g., tumor, infection, active demyelinating lesion), MRI with gadolinium is also indicated.

It is important to note that the metabolic hyperactivity and changes in blood flow caused by seizures can induce transient, variably reversible changes on numerous MRI sequences (most notably on T2, FLAIR, DWI-ADC, and gadolinium-enhanced images) which may be widespread and lead to a broad, and potentially misleading, differential diagnosis [29].

Dynamic imaging with PET and single photon emission computed tomography (SPECT) is emerging as a useful tool in the evaluation of the metabolic effects of FSE-EPC, especially when MRI is normal [22]. Because of the relative technical ease of obtaining a SPECT in patients having continuous focal seizures, this test can be particularly useful in the evaluation of EPC [30]. It can be used to clarify confusing situations when EPC is suspected but EEG fails to show epileptic changes.

EPC is notoriously refractory to medical treatment [5]. As with all forms of epilepsy, it is essential to identify the underlying etiology and, when possible, treat appropriately for the underlying disease. For example, because nonketotic hyperglycemia and resultant hyponatremia are such frequent causes of EPC, it is crucial to obtain a complete metabolic panel and treat the underlying metabolic derangement. In patients with specific autoimmune diseases like Rasmussen encephalitis, treatment of the underlying disease process is key.

Anti-seizure drugs (ASDs) are the mainstay in the initial treatment of EPC, but their efficacy is usually limited and often, multiple ASDs are required to achieve a sustained effect [5]. In general, ASDs help to prevent the spread of EPC into complex partial or secondarily generalized seizures, but rarely do they appear to alter the severity of the EPC significantly. As in other types of SE, benzodiazepines are the most effective first-line drug for urgent interruption of EPC [31], but complete suppression often necessitates prohibitively high doses that may lead to marked sedation and respiratory depression.

There have been no large or randomized trials evaluating the efficacy of particular ASDs in the treatment of EPC. A retrospective, multicenter study of 65 cases of EPC (excluding patients with acute stroke or Rasmussen encephalitis as the etiology) found that topiramate and levetiracetam yielded a better overall success rate [31]. Topiramate was effective in 7 of 28 cases in which it was tried, and patients with a dysontogenetic etiology (four cortical dysplasias and one arteriovenous malformation) appeared to respond best. Levetiracetam was given to 26 patients and was successful in 8, five of whom had inflammatory etiologies. Other ASDs including valproic acid and lacosamide have been reported as safe and effective in individual case reports, but in the absence of clinical trial data, we recommend that the choice of ASD be tailored to the individual case and side effect profile.

All patients presenting with FSE should be assessed carefully for an underlying lesion that may be amenable to curative resective surgery. Currently, the surgical literature for treatment of FSE is limited to case reports and case series. The majority of patients had EPC. The most common operations were focal resection, lobar or multilobar resections, hemispherectomy (functional, anatomical, or modified), and corpus callosotomy [24, 32]. Other surgical treatments have included multiple subpial transections, implantation of a vagus nerve stimulator, low-frequency repetitive cortical electrical stimulation, and thalamic deep brain stimulation [32–34].

A recent compilation of 23 case reports of the surgical treatment of FSE found that seizure freedom was achieved in 18 of the 23 patients, with follow-up periods of 4 months to 5 years [24]. A minority of patients had continuing but improved seizures at the time of reporting, without worsening of seizure frequency in any. Of note, the majority of the patients operated on were young (ages 8 days to 36 years), and had strong semiologic, imaging, or electrographic evidence (or combinations of these) of focal epilepsy, usually presenting with refractory status epilepticus on a background of habitual seizures. The great majority of patients had focal or hemispheric malformations of cortical development as the etiology.

One of the larger surgical case series included in the meta-analysis above involved 10 children with FSE refractory to high-dose ASDs who had various surgical treatments including callosotomy, lobectomy, or anatomic and functional hemispherectomy [31]. Their underlying illnesses included malformations of cortical development (n = 6), tuberous sclerosis (n = 1), Rasmussen encephalitis (n = 1), prenatal large-artery infarct (n = 1), and an unclear diagnosis (n = 1). In this study, SE was stopped by surgery in 100% of patients, with no perioperative mortality and with significant postoperative improvement in functional status.

Thus, based on the limited current literature, patients with convulsive or nonconvulsive refractory status epilepticus who have a high degree of concordance among semiology, imaging, functional imaging with PET/SPECT, and EEG (scalp as well as invasive) indicating a similar single epileptogenic zone (with focal cortical dysplasia as the underlying pathology) appear most likely to benefit from surgery. In patients with a nonlesional MRI or a poorly defined epileptogenic zone or both, invasive EEG monitoring should be considered strongly [24].

Unfortunately, the majority of patients with FSE are not candidates for surgery, for various reasons, e.g., when the expected motor deficit resulting from the removal of motor cortex is considered unacceptable, when the patient is otherwise neurologically normal, when EPC is bilateral, or when the case is nonlesional. A fascinating report describes a case of “mirror EPC,” in which resection of a focal cortical dysplasia causing EPC resulted in the development of contralateral EPC in a previously undetected region of cortical dysplasia in the opposite hemisphere [35].

Fortunately, noninvasive techniques for the treatment of FSE are now being developed. A case report and case series found repetitive transcranial magnetic stimulation (rTMS) safe and potentially effective in the treatment of EPC [36, 37]. rTMS is a noninvasive method for focal cortical stimulation during which small intracranial electrical currents are generated repeatedly and applied by a strong fluctuating extracranial magnetic field. Interictal rTMS delivered over a neocortical seizure focus has been demonstrated to reduce seizure frequency in some patients, reducing cortical excitability via a yet unknown mechanism. Ictal rTMS is less well studied. The largest case series of ictal rTMS included seven patients with EPC of mixed etiologies. rTMS resulted in a brief (20–30 min) pause in seizures in three patients and a lasting (>1 day) pause in two more; the other two had no interruption in seizures [37]. Seizures were not exacerbated by rTMS in any patient, and side effects were generally mild (e.g., transient head or limb pain, or limb stiffening during high-frequency rTMS trains). Larger and more controlled studies are needed to explore the utility of rTMS in FSE, but at this time it is one of the only noninvasive, non-pharmacologic options for treatment.