The mortality in a cohort of 472 children (aged 1 month to 16 years) with epilepsy, newly diagnosed between 1988 and 1992, was compared in a multicenter study with that in the same age group in the general population in the Netherlands. All children were followed for 5 years or until death. Nine children died, and the mortality rate was 3.8/1000 person-years, seven times higher than expected. Death occurred only among children with epilepsy caused by a static or progressive neurologic disorder (6 with neonatal encephalopathy, 3 progressive disorder, 1 Niemann-Pick C, 1 ependymoma, and 1 infantile ceroid lipofuscinosis). None had a sudden unexpected and unexplained death from epilepsy. No mortality occurred among 328 children with epilepsy of nonsymptyomatic cause. In those with symptomatic epilepsy, the mortality risk was 22.9 vs 0.39 expected. 
COMMENT. Compared to the general pediatric population in the Netherlands, the risk of mortality in childhood epilepsy is not increased for nonsymptomatic cases but reaches a 20-fold increase in children with symptomatic epilepsy. Among the 9 deaths in this prospective study, none was sudden unexpected and unexplained (SUDEP). An accurate initial diagnosis is important in counseling parents regarding risks of mortality in epilepsy.
Dr Neil Gordon, Wilmslow, Cheshire, UK, reviews SUDEP in the May issue of . In SUDEP, by definition, death is not the immediate result of a seizure or status epilepticus. In one report , 91% of 57 with SUDEP had undergone autopsies; risk of SUDEP was not increased in patients with symptomatic as opposed to idiopathic epilepsy. Risk factors included early-onset epilepsy, poor seizure control, polytherapy with AEDs, and frequent dose adjustments or abrupt AED withdrawal. Causes of SUDEP are usually multiple. Difficulties in establishing cause using data from death certificates is stressed by Appleton RE.  who reported findings in 60 children with SUDEP.
In a study reviewed in Ped Neur Briefs (March 2001; 15:24), the mortality rate of children in antiepileptic drug trials was 4.1 per 1000 person years, and the SUDEP rate was 2.4/1000 person years. Only age was associated with the risk of SUDEP, and disease severity is the probable determining factor. Length of epilepsy history, gender, and number of concomitant drugs do not influence the SUDEP rate . Also, see , for further recent SUDEP study.