Mortality in epilepsy : epidemiological studies with emphasis on sudden unexpected death and suicide

Abstract

The general aim of this study was to create, through an increase in knowledge, the preconditions for reducing epilepsy patients' significantly increased mortality The main aim was to identify clinical characteristics of patients in a large cohort of adult patients with a diagnosis of epilepsy who had a) undergone sudden unexpected death in epilepsy (SUDEP), or b) committed suicide. Another aim of this study was to describe overall and cause specific mortality in this cohort. We have also studied overall and SUDEP mortality in Swedish epilepsy surgery (ES) patients.

9061 patients were present in the Stockholm County In-Ward Register who at least once during 1980-89 at 15 years of age or older had been hospitalized and discharged with a diagnosis of epilepsy. They were followed in the Swedish Cause of Death Register until the end of 1992. 4001 patients had died, the standardized mortality ratio (SMR) for all causes of death was 3.6 (3.5-3.7 95% Cl). SMR was significantly increased for all ages and in a wide range of causes of death, showing that this large subgroup of patients with a diagnosis of epilepsy is a population at risk.

57 cases of SUDEP were compared with 171 controls, living patients from the study cohort. We found strong associations between the relative risk (RR) of SUDEP and early onset of epilepsy, high seizure frequency, polytherapy with antiepileptic drugs (AED) and frequent changes of dosages of AEDs. Adjustment for seizure frequency revealed that these variables each contribute to the risk and not only act as surrogates for the severity of the seizure disorder. There were weaker associations with lack of therapeutic drug monitoring (TDM) and plasma concentrations of carbamazepine above the usually recommended interval. We found no increase in RR for SUDEP with gender, type of epilepsy, localization of epileptic focus, alcohol abuse or with the use of any particular AED.

26 cases of suicide were compared with 171 living control patients from the study cohort. We found strong associations between the risk of suicide and psychiatric illness. The only epilepsy-related risk factors for suicide were signs of inadequate neurologic follow-up and onset of epilepsy in childhood or adolescence compared with onset at higher ages.

All patients who had epilepsy surgery (ES) treatment during 1990-98 (N=596) were identified through the Swedish Epilepsy Surgery Register and followed in the cause of death register until the end of 1998. The SMR for all causes of death was 4.9 (2.7-8.3 95% Cl) and the incidence of SUDEP was 2.4/1000 person years, i.e. in the same range as in other similar cohorts. We could not demonstrate any differences in mortality by type of operation, localization of resection, or seizure situation at 2-year postoperative follow-op. None of the SUDEP patients were seizure free by the time of death. The mortality and SUDEP incidence seemed to be higher for patients rejected for surgery than for surgery patients.