Abstracts

Rationale: Treatment for prolonged refractory status epilepticus (incomplete seizure control despite anesthetic infusion for >7 days; PRSE) is based on minimal data.Methods: Three academic centers within the Critical Care EEG Monitoring Research Consortium systematically reviewed records to identify patients with PRSE. Modified Rankin Score (MRS) at discharge and most recent follow-up was noted.Results: Patient population: 63 patients were identified with PRSE (excluding cardiopulmonary arrest). The most common AEDs used were phenytoin (95%), levetiracetam (92%), lorazepam (60%), phenobarbital (59%) and valproate (54%). The most common infusions were pentobarbital (81%), propofol (76%) and midazolam (70%). In most cases, no particular therapy was credited with permanent SE cessation. Non-traditional therapy: Ketamine was used in nine patients (13%); one patient also received it orally. While ketamine was never credited with permanent cessation of status epilepticus (SE), two patients survived with recent MRS of 2-3. Surgical resection for seizure control was done in four; one had meningioma, two previously had CNS tumors excised (pilocytic astrocytoma and metastasis) and one had focal epilepsy. Surgery terminated SE in 3/4 (all but meningioma). Two survived with current MRS of 2-3. Therapeutic hypothermia was used in three patients, though it did not terminate SE for any; one survived. Steroids were used in 13 patients, of which two had NMDA receptor-associated encephalitis, one epilepsy, two presumed vasculitis, two an inflammatory etiology, one meningioma and five presumed CNS infection (no identified organism). Steroids were credited with permanent cessation of SE in one (CNS vasculitis). IVIG was administered to six patients (two with NMDA receptor-associated encephalitis, two inflammatory/autoimmune etiology and two presumed viral encephalitis) and was successful in one (inflammatory disease). Plasmapheresis was used in all three cases of NMDA receptor-associated encephalitis, with 2/3 returning to their baseline and one remaining in a persistent vegetative state (PVS). In no case was plasmapheresis considered responsible for SE cessation. All three patients with NMDA receptor-associated encephalitis had bilateral oophorectomies and 2/3 received rituximab; neither therapy was considered effective on its own. One patient with epilepsy received electroconvulsive therapy which was ineffective. One patient with presumed viral encephalitis was given a ketogenic diet; it was ineffective and that patient remains in a PVS. Other medications used included lidocaine in two patients and bromide in one. Of 23 patients who received non-traditional therapies for PRSE, 8 had favorable outcomes with most recent MRS of 1 (n=4), 2 (n=3) and 3 (n=1). Five patients had a MRS of 5 at last follow-up.Conclusions: Non-traditional therapies and alternative medications may be used for PRSE, and patients can sometimes recover close to full function. However, none of these was consistently (or even usually) effective. Additional, large multicenter studies are needed.