By Padmaja Kandula, MD
Assistant Professor of Neurology and Neuroscience, Comprehensive Epilepsy Center, Weill Cornell Medical College

Dr. Kandula reports no financial relationships relevant to this field of study.

Synopsis: In a combined derivation and validation study, three independent risk factors for refractory status epilepticus were identified — acute symptomatic cause for seizures, stupor or coma, and a low serum albumin < 35 g/L.

Source: Sutter R, et al; Early predictors of refractory status epilepticus: An international two-center study. Eur J Neurol 2015;22:79-85.

Over the years, the consensus definition of refractory status epilepticus has evolved into the persistence of the ictal state after failure of first- and second-line antiepileptic agents. However, the aggressiveness of treatment is still under fierce debate. The potential risk-to-benefit ratio of anesthetic treatment, along with limited technical resources at many hospitals, continues to plague practitioners. In this study, Sutter et al tried to predict refractory status epilepticus by applying available clinical information to a matched European derivation data set and a U.S. validation set.

The authors identified 171 patients from the Swiss derivation data set and 131 patients from a U.S. validation set from 2005-2012 after exclusion of post-anoxic status epilepticus. Status epilepticus was defined as clinical and electroencephalographic seizure activity lasting at least 5 minutes or seizures without interval recovery. Etiology (acute vs non-acute), level of consciousness (awake/somnolent vs stuporous/comatose), and serum albumin levels were recorded in all patients. All included patients underwent continuous electroencephalography (cEEG) or samples of > 20 minutes of EEG every 12 hours, where continuous EEG was not available. A uniform treatment protocol with initial benzodiazepine treatment followed by intravenous phenytoin/fosphenytoin, valproic acid, levetiracetam, or combination thereof was standardized for all patients. After failure of the above first- and second-line agents, anesthetic coma was subsequently induced. The primary outcome (refractory status) was defined as clinical and/or electrographic status persistence despite first- and second-line therapy.

The percentage of patients with refractory status epilepticus was 45% in the derivation set and 46.6% in the validation set. Acute status epilepticus etiology, coma/stupor, and serum albumin < 35 g/L at status onset were independent predictors for refractory status in the derivation data set (odds ratios = 2.02, 4.83, and 2.45, respectively, for the three clinical conditions). These results were externally validated with an independent U.S. validation set.


Although retrospective in nature, this study by Sutter and colleagues identified three useful, readily available clinical parameters to assess potential refractory status epilepticus. The use of two large cohorts also allows a good prediction and validation model. On the other hand, the relatively younger average age (66-67 years) of both cohorts and exclusion of post-anoxic status epilepticus limits generalization of these results. In addition, cases in which only periodic routine EEG was available may have missed subclinical or electrographic status with a subsequent bias toward a more favorable outcome. A further prospective trial with larger sample size to allow adequate status etiology sub-typing (including anoxic status) has the potential to be useful in prognostication and allocation of resources influencing aggressiveness of overall treatment.