Abstracts

Patients with epilepsy have increased mortality compared to the general population. Most of this excess mortality is attributable to the underlying cause of epilepsy such as brain tumours; however seizures themselves are also believed to cause death. Whether risk of death should be discussed with patients at the time of diagnosis is a controversial issue. We analysed mortality data in a cohort of 890 patients with newly diagnosed epilepsy to quantify the risk of excess mortality. All patients diagnosed with epilepsy at the Epilepsy Unit, Western Infirmary, Glasgow between July 1981 and May 2001 were included in the study. The patient list was matched against the Death Database at the General Register Office for Scotland in August 2002. Death certificate copies were obtained for patients whose demographic information matched. All surviving patients were prospectively flagged on the National Health Service Central Register. All deaths reported till September 2003 were included in the analysis. Comparison was made with an age- and sex- matched cohort constructed from the UK Government Actuary[apos]s Departments Interim Life Tables to reflect the expected mortality in the Scottish population. Expected number of deaths was calculated by summing individual expected mortality over the study period. Observed and expected deaths were compared using a [chi]² statistic on 1 degree of freedom. Overall, 93 deaths were observed in the cohort (crude death rate 10.4%), compared 65.54 deaths expected (Rate ratio 1.41 95% CI 1.15-1.74). Kaplan Meier method with log-rank test was used to compare survival in patients with idiopathic generalised, symptomatic partial and cryptogenic partial epilepsies. Symptomatic epilepsy had the worst survival prognosis compared to both idiopathic and cryptogenic epilepsies (p[lt]0.0001). Age- and sex-specific standardised mortality ratios (SMRs) were calculated for the cohort to quantify excess mortality. The highest SMRs were seen for patients aged less than 40 years at diagnosis (SMR 4.25 95% CI 2.70-6.30, p[lt]0.0001). This was especially high in patients diagnosed with symptomatic epilepsy under the age of 40 (SMR 8.64 95% CI 3.78-13.50, p[lt]0.0001). There was significant excess mortality in patients aged [lt]40 with idiopathic generalised epilepsy (SMR 4.62. 2.01-8.94), as the expected mortality was low in this predominantly younger group of patients. Patients diagnosed with epilepsy are at significantly higher risk of dying compared to age and sex matched controls. This excess mortality is highest in younger patients, especially those with symptomatic epilepsy. Higher number of deaths was also observed in younger patients with idiopathic generalised epilepsy, suggesting that seizures themselves contribute to mortality. These factors should be taken into account while counselling patients diagnosed with epilepsy.