Risk factors for sudden unexpected death in epilepsy: a case-control study

Background Sudden unexpected death is substantially more common in people w ith epilepsy than in the general population. Our objective was to investiga te the association between some clinical variables and sudden unexpected de ath in epilepsy (SUDEP) to identify risk factors. Methods This nested case-control study was based on a Cohort of people aged between 15 and 70 years, who, during 1980-89, had been admitted to and dis charged with a diagnosis of epilepsy from any hospital in the county of Sto ckholm. The study population was followed up through the National Cause of Death Register until Dec 31, 1991. Cases were individuals who had died, wit h a diagnosis of epilepsy registered on the death certificate, and who afte r review of medical and necropsy records were found to meet our SUDEP crite ria. Three control participants, who were living epilepsy patients matched for age and sex, were selected from the same cohort for each case. All medi cal records were examined. Clinical data were collected and analysed on a p redesigned protocol. Findings 57 SUDEP cases were included, of whom 91% had undergone necropsy. The relative risk of SUDEP increased with number of seizures per year. The estimated relative risk was 10.16 (95% CI 2.94-35.18) in patients with more than 50 seizures per year, compared with those with up to two seizures per year, The risk of SUDEP increased with increasing number of antiepileptic drugs takenconcomitantly-9.89 (3.20-30.60) for three antiepileptic drugs co mpared with monotherapy. Other major risk factors were early-onset versus l ate-onset epilepsy (7.72 [2.13-27.96]), and frequent changes of antiepilept ic drug dosage compared with unchanged dosage (6.08 [1.99-18.56]). The asso ciation between SUDEP risk and early onset, and SUDEP risk and seizure freq uency, was weaker for female than for male patients, whereas frequent dose changes showed a stronger association in female patients. Interpretation Our data suggest that SUDEP is a seizure-related event, alth ough the pathophysiological substrate that predisposes individuals to SUDEP may be established at an early age, and there may be some sex differences. Improvement of seizure control and possibly the avoidance of polytherapy m ay be ways to reduce the risk of SUDEP.