Abstracts

Rationale: While many patients who are assessed for potential new-onset seizures are diagnosed as having had an epileptic seizure, a significant proportion will be diagnosed with a non-epileptic event. Little is known about specific mortality risk for these different patient groups. This study aimed to examine mortality risk for major diagnostic groups within a cohort of patients who presented for assessment following a suspected new-onset of epileptic seizure. Methods: The cohort (N=993) comprised patients without a prior diagnosis of epilepsy who were assessed from 2000-2009 in the adult First Seizure Clinic at The Royal Melbourne Hospital, Australia. Linkage with National Death Index identified deceased patients and cause of death. Patient deaths were compared to Australian age and sex specific rates using Standardized Mortality Ratios (SMR). Results: Median time post-diagnosis was 4.5 years (range 0.1-11). Among patients diagnosed with an epileptic seizure (n=713), 67 patients had deceased. Patients had an overall rate of death that was more than double that expected compared to population rates (SMR 2.2; 95%CI 1.7-2.7). The risk remained consistently elevated for at least 10 years post-diagnosis (years 0-1 post-diagnosis SMR 2.2, years 1-5 SMR 2.0, years 5-10 SMR 2.5). Epilepsy was the primary cause of death in only 7/67 deaths. One additional patient died from a brain tumor and the cause was unknown in 9 cases. Amongst the 180 patients diagnosed with a non-epileptic event (114 syncopal, 66 psychogenic), ten patients had deceased with an overall patient mortality > twice the population rate (SMR 2.5, 95% CI 1.3-4.6). In a further 100 patients diagnosis was uncertain. Of these 5 had deceased, mortality did not differ from the population (SMR 1.2, 95% CI 0.5-2.8). Conclusions: Patients with a new diagnosis of epilepsy demonstrated consistently elevated SMRs up to 10 years post-presentation. Most deaths were not directly related to epilepsy, therefore counselling and reinforcement of general preventative health messages may be appropriate for this group. Of the other patient groups assessed for suspected new-onset seizures, those ultimately diagnosed with non-epileptic events were also at increased risk of mortality. These events may therefore be early indicators of other disorders and routine referral for further assessment and follow-up of these patients may be warranted.