Sudden unexpected death in epilepsy, incidence, circumstances and risk factors

Abstract: Although Sudden Unexpected Death in Epilepsy (SUDEP) has attracted increasing attention from the scientific community during the last 20 years, important gaps in knowledge still exist that hamper the development of methods aiming at prevention of this, the most devastating consequence of epilepsy. We are still missing large population-based studies on the incidence of SUDEP. Our understanding of the circumstances surrounding SUDEP is incomplete which is a major limitation when it comes to development of potential SUDEP-preventing devices. Finally, our understanding of risk factors for SUDEP is limited to a few established risk factors. The purpose of this study was to examine the incidence, circumstances and risk factors for SUDEP in Sweden. The project is based on a study population (n=78 524) which comprises all persons living in Sweden at 1. July 2006, who at some point during 1998-2005 where registered with the diagnosis code for epilepsy (ICD G 40) in the Swedish National Patient Register (SNPR). To identify cases of SUDEP, the study population was linked to the National Cause-of- Death Register. During the follow-up time from July 1, 2006 to December 31, 2011, we identified 9605 deaths. All death certificates in the study population between 1 July 2006 and 31 December 2011 with epilepsy mentioned on death certificate and all deaths during 2008 (n=3166) were reviewed. Based on the information in the death certificates, obvious non-SUDEP deaths were excluded from further analysis. For all others we analyzed patient medical records, autopsy and police reports and information was extracted using a standardized protocol. From the study population, five epilepsy controls per SUDEP case, of the same sex, who were alive at the case´s time of death, were randomly selected by the National Board of Health and Welfare. During 2008, 1890 individuals from the study population died. Of these, 99 met Annegers‘ SUDEP criteria (49 definite, 19 probable, and 31 possible) (paper I). Definite and probable SUDEP accounted for 3.6% of all deaths in the study population during 2008, and 5.2% when possible was included. In the age group 0-15 years, the relative contribution of SUDEP (definite, probable and possible) to overall deaths was 36.0%. SUDEP incidence was 1.20/1000 person-years (definite/ probable) and 1.74/1000 if possible SUDEP was included. Epilepsy was mentioned in any position of the death certificate in 63.6% of the 99 SUDEP cases. Of the 329 SUDEP deaths identified from July 1, 2006 to December 31, 2011 (167 definite, 89 probable, 73 possible), more than half (58%) occurred at night and 91% died at home, whereof 65% were found deceased in bed (paper II). Death was witnessed in 17% of all SUDEP cases and when a seizure was witnessed in conjunction with SUDEP (n=49) all were generalized tonic-clonic seizures (GTCS). Where a body position was documented (43%), more than two thirds (70%) were found prone. Dying at night made it more likely (80%) to be found prone than other times (55%) (p<0.001). Among adult SUDEP cases, 75% were living alone, and only 14% of all SUDEP cases shared a bedroom. In papers III and IV, 255 SUDEP cases (167 definite, 88 probable) were compared to their matched 1148 controls. Those with a history of GTCS had a tenfold increased SUDEP risk and the risk was increased to 32-fold with 4-10 GTCS during the last year of observation. When a history of nocturnal GTCS was present, a nine-fold SUDEP risk was observed and a 15-fold risk was seen if nocturnal GTCS were present during the last year of observation. No increased risk of SUDEP was seen in those experiencing exclusively non-GTCS during the preceding year. There was a fivefold increased risk of SUDEP among those living alone, while the risk was reduced to twofold when sharing household but not bedroom. Individuals experiencing ≥1 GTCS and not sharing a bedroom with someone had 67-fold increased risk of SUDEP compared to individuals not having GTCS, who shared their bedroom with someone, with attributable proportion due to interaction estimated at 0.69 (95% confidence interval, CI 0.53-0.85). Polytherapy, especially taking three or more AEDs was associated with a 69% reduced SUDEP risk after adjusting for GTCS frequency and other covariates. Levetiracetam as monotherapy was associated with a significantly lower SUDEP risk when compared to no AED treatment (odds ratio, OR 0.10, 95% CI 0.03-0.61). Lamotrigine, valproic acid and levetiracetam were associated with a significantly reduced risk when used as part of a polytherapy. Use of statins was associated with a reduced risk of SUDEP (OR 0.34, 95% CI 0.11-0.99). Our results show that SUDEP is an important contributor to mortality in epilepsy patients, and accounts for one third of deaths in children with epilepsy and one fifth of deaths among young adults with epilepsy. Since the majority died at home in bed, at night with indications of a previous GTCS, SUDEP can be considered an event related to night time and unobserved GTCS. We found no excess risk of SUDEP among individuals experiencing non-GTCS only, which has important clinical implications. GTCS and lack of supervision were the main risk factors. Moreover, our results suggest that up to 69% of SUDEP cases could be prevented in individuals with GTCS who live alone, if they were made free from GTCS or did not sleep alone. Polytherapy was associated with a substantially reduced SUDEP risk indicating that physicians should to consider AED polytherapy more pro-actively for patients with poorly controlled GTCS.