Most patients with epilepsy achieve seizure control with medications. However, about 30% of patients have drug-resistant epilepsy. Drug-resistant epilepsy, often known as medically intractable epilepsy, is defined as continued seizures despite adequate trials of two tolerated, appropriately chosen and dosed antiepileptic drugs (either as monotherapy or in combination). Early referral to an epilepsy center is recommended, as patients with drug-resistant epilepsy have higher rates of disability, injuries, and even death compared to those with well-controlled epilepsy.
Prior to categorizing epilepsy as drug-resistant, so-called “pseudo-resistance” should be excluded. This requires ensuring that (1) epilepsy is the correct diagnosis (ruling out syncope, sleep disorders, cardiac arrhythmias, and psychogenic nonepileptic seizures), (2) the selected antiepileptic medication is appropriate to the epilepsy type (generalized versus focal) and given at an adequate dose, and (3) poor or variable adherence to medication or provoking factors such as substance abuse are not thought to be the primary cause of continued seizures. Of note, intolerable side effects from an antiepileptic medication do not necessarily represent failure of efficacy, so more than two agents may need to be tried if the initial agents are limited by tolerability.
Patients with an identifiable seizure focus are potential candidates for resective surgery, which is more likely than continued medical therapy to lead to seizure freedom. Seizure remission rates are highest in cases when the lesion seen on magnetic resonance imaging (MRI) correlates with the anatomic focus identified on electroencephalogram (EEG). Counseling about the risk of sudden unexpected death in epilepsy (SUDEP), surgical risks, medication side effects, and cognitive and psychiatric morbidity associated with either surgical treatment or with continued uncontrolled seizures should be provided to the patient. The initial evaluation typically includes at least video EEG, high-resolution brain MRI, neuropsychological testing, and positron emission tomography (PET) scan. More extensive testing with ictal single photon emission computed tomography (SPECT), magnetoencephalography (MEG), functional MRI (fMRI), or other advanced imaging may be necessary as well.
If the epileptogenic zone (the optimal target for surgical resection to achieve seizure freedom) remains undefined after initial evaluation, more invasive monitoring can be considered. Intracranial EEG with electrode grids, strips, or stereotactically placed depth electrodes can aid in localization to identify the seizure focus in many cases. Intracranial EEG can also help to evaluate the proximity of seizure foci to “eloquent” cortex, to limit the risk of postoperative neurologic deficits.
If potentially curative resection or ablation are not considered good options due to inability to identify a single resectable focus or concern that resection would lead to unacceptable neurologic deficits, palliative options should be considered. Palliative surgery with incomplete resection of the epileptogenic zone may still provide significant benefit by decreasing the frequency or severity of seizures. Implanted devices including vagus nerve stimulation (VNS), responsive neurostimulation (RNS), and deep brain stimulation (DBS) to the anterior nucleus of the thalamus are generally not curative but do decrease seizure frequency, and may provide additional benefits in terms of mood or cognition.
For patients with drug-resistant generalized, or mixed, epilepsies, VNS, DBS, and a ketogenic diet, or other dietary therapy, may provide benefit. Corpus callostomy can prevent seizures starting in one hemisphere from spreading bilaterally and can thereby help prevent drop attacks such as those seen in Lennox-Gastaut syndrome.