Article Text


Non-convulsive status epilepticus
  1. D Audenino1,
  2. L Cocito1,
  3. A Primavera1
  1. 1Department of Neurosciences, University of Genova, via De Toni 5, Genova 16132, Italy
  1. Correspondence to:
 Dr Daniela Audenino;
  1. A M Husain2
  1. 2Box 3678, 202 Bell Building, Duke University Medical Center, Durham NC 27707, USA;

    Statistics from

    In a recent article, Husain et al argue that severely impaired mental state, ocular movement abnormalities, and the patient’s history could suggest a diagnosis of non-convulsive status epilepticus (NCSE), and then be a selection criterion for patients with impaired consciousness to undergo an urgent EEG recording.1 However, our experience with 50 adult patients (12 men and 38 women, mean age 65.9 years) meeting the criteria for the diagnosis of NCSE2 suggests that there are no peculiar clinical features characteristic of this condition. Twenty eight patients had absence status, 16 had complex partial status, and six had aphasic status; 11 had tonic-clonic seizures just before the onset of status, and eight had a history of chronic epilepsy. The main clinical presentation of NCSE was aphasia (six patients), mutacism (two patients), psychiatric disturbances (four patients), delirium (34 patients), and stupor or coma (four patients). We feel therefore that all acute alterations of mental state or changes in behaviour from baseline for which no alternative explanation is available may raise a suspicion of NCSE and call for urgent EEG. In patients with known epilepsy presenting with prolonged confusion after convulsions, an urgent EEG is warranted to distinguish between postictal encephalopathy and generalised NCSE.

    Kaplan3 reported that the diagnosis of NCSE was initially missed in the emergency room when the behavioural or cognitive changes from baseline were ascribed to other causes, including intoxication, postictal encephalopathy, pre-existing psychiatric conditions, or mental retardation. In our study, the delay in diagnosis of NCSE ranged from three hours to 28 days, with a mean of three days. In two patients the diagnosis was delayed despite an emergency EEG, because of the lack of clear epileptiform features in the EEG abnormalities. A suspicion of NCSE is the most important clinical indication for performing an emergency EEG.4 Because of the different EEG patterns and the pleomorphic clinical features, a diagnosis of NCSE is only possible with an expert integration of EEG findings and clinical data, as emphasised by Niedermeyer and Ribeiro.5


    Author’s reply

    The interest and comments of Audenino et al are greatly appreciated. In our paper, 48 patients who were suspected of being in NCSE were evaluated prospectively by neurology residents; the diagnosis of NCSE was later confirmed or ruled out on the basis of the patient’s EEG. Remote risk factors for seizures (such as previous stroke, neurosurgery, significant head trauma), impaired mental status, and ocular movement abnormalities (sustained eye deviation, nystagmus, hippus) were found significantly more often in the NCSE group. The combined sensitivity of remote risk factors for seizures and ocular movement abnormalities was 100%; there was no patient in the NCSE group who did not have either of these findings.

    Audenino et al present their series of 50 patients, all of whom met their criteria for NCSE. In their series, like ours, women outnumbered men. They note that there are “no peculiar clinical features of NCSE.” They also suggest that all alterations of behaviour for which there is no alternative explanation should be evaluated with an urgent EEG.

    I agree that for all patients with an altered sensorium, NCSE should be considered as a possible diagnosis. Furthermore, there is general agreement that an EEG is required for the diagnosis of NCSE. In our study it was also noted that the presence of a metabolic or other type of encephalopathy did not necessarily imply exclusion of NCSE. Thus should urgent EEG be requested for all patients with altered mental status, regardless of comorbidities? No; this is impractical not only after working hours, but also during working hours in most hospitals. Therefore, we attempted to triage patients who should be getting an urgent EEG. This can be accomplished on the basis of the high sensitivity of the above findings. This high sensitivity should not be mistaken for high specificity; in fact the specificity was low. A valid criticism would be the low specificity, but our objective was not to find an alternative to EEG for the diagnosis of NCSE, but rather to triage those in need of one.

    Another important difference between Audenino’s series and our own is that ours was obtained prospectively. A prospectively obtained neurological history and examination is likely to be more detailed than information obtained retrospectively from a review of the clinical records. A history of remote risk factors for seizures and the presence of ocular abnormalities can be missed during an urgent neurological evaluation unless specifically sought.

    I would like to emphasise again that the objective of this study was not to find alternatives to EEG in the diagnosis of NCSE, but rather to help select those who should have an urgent EEG. Performing an urgent EEG on every patient with alterations in mental status is not practical or possible in most institutions.

    View Abstract

    Request permissions

    If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.