No abstract

No abstract

SUMMARYSudden unexpected death in epilepsy (SUDEP) is a category of death in people with epilepsy occurring in the absence of a known structural cause of death and is most likely heterogeneous with regard to mechanisms and circumstances. SUDEP is particularly difficult to investigate in research studies for several reasons, including its relatively low incidence, its unpredictable occurrence often in unwitnessed settings, and its low rate of complete autopsy examinations. Over the past two decades, two complementary definitions have been used in most SUDEP studies, but often with variations. We propose here a unified SUDEP definition and classification to resolve current ambiguities and to retrieve cases that would not have been further studied if the previous definitions were used. The proposed Unified SUDEP Definition and Classification contains, in addition to concepts inherent in the previous definitions, nine main recommendations. (1) The word ''unexpected,'' and not the word ''unexplained,'' should be uniformly used in the term SUDEP. (2) The SUDEP category should be applied when appropriate, whether or not a terminal seizure is known to have occurred. (3) The ''Possible SUDEP'' category should be used only for cases with competing causes of death, with cases left unclassified when data are insufficient to reasonably permit their classification. (4) Cases that would otherwise fulfill the definition of SUDEP should be designated as ''SUDEP Plus'' when evidence indicates that a preexisting condition, known before or after autopsy, could have contributed to the death, which otherwise is classified as SUDEP (e.g., coronary insufficiency with no evidence of myocardial infarction or long-QT syndrome with no documented primary ventricular arrhythmia leading to death). (5) To be considered SUDEP, the death should have occurred within 1 h from the onset of a known terminal event. (6) For status epilepticus as an exclusion criterion for SUDEP, the duration of seizure activity should be 30 min or more. (7) A specific category of SU-DEP due to asphyxia should not be designated, the distinction being largely impractical on circumstantial or autopsy evidence, with more than one mechanism likely to be contributory in many cases. (8) Death occurring in water but without circumstantial or autopsy evidence of submersion should be classified as ''Possible SUDEP.'' If any evidence of submersion is present, the death should not be classified as SUDEP. (9) A category of ''Near-SUDEP'' should be agreed to include cases in which cardiorespiratory arrest was reversed by resuscitation efforts with subsequent survival for more than 1 h. Scenarios that demonstrate the basis for each SUDEP category are described. If disagreement exists about which category fits a particular case, we suggest the use of consensus decision by a panel of informed reviewers to adjudicate the classification of the case.

This work lends support to the view that SUDEP is a seizure-related phenomenon and that control of tonic-clonic seizures is important in its prevention. Nocturnal supervision seems to protect against SUDEP.

Objective-To record non-invasively ictal cardiorespiratory variables. Methods-Techniques employed in polysomnography were used in patients with epilepsy undergoing EEG-video recording at a telemetry unit. Results-Apnoea (> 10, range > 10-63, mean 24 s) was seen in 20 of 47 clinical seizures (three secondary generalised, 16 complex partial, and one tonic) and 10 of 17 patients. Apnoea was central in 10 patients, but obstructive apnoea was also recorded in three of 10. Oxyhaemoglobin saturation (Spo,) dropped to less than 85% in 10 seizures (six patients). An increase in heart rate was common (91% of seizures). Bradycardia/sinus arrest was documented in four patients (mean maximum RR interval 5 36, range 2-8-8-6 s) but always in the context of a change in respiratory pattern. Conclusion-Ictal apnoea was often seen. The occurrence ofbradycardia in association with apnoea suggests the involvement of cardiorespiratory reflexes. Similar mechanisms may operate in cases of sudden death in epilepsy.