Mortality in Epilepsy
Mortality in Epilepsy
ABSTRACT & COMMENTARY
Source: Nilsson L, et al. Cause-specific mortality in epilepsy: A cohort study of more than 9000 patients once hospitalized for epilepsy. Epilepsia 1997;38:1062-1068.
Most research on epilepsy-associated mortality has studied relatively small institutionalized or referral center populations with severe seizure disorders. Nilsson et al now describe mortality in a large community-based epilepsy cohort.
Using the Stockholm County In-Patient register, the authors identified and studied all patients older than 14 years who were discharged from any hospital in their county from 1980 to 1989 with a diagnosis of epilepsy. Each patient was studied until death or until December 31, 1992. The cause-of-death data were derived from the Swedish National Cause-of-Death Register. This Register is created from death certificates. In 48% of cases, information on the death certificates was based on autopsy reports.
The observed number of deaths in this epilepsy cohort was compared with the number of deaths in the general population of Stockholm County over the same time period. Standardized mortality ratios (SMRs), which are the ratios of observed deaths to expected deaths, were calculated.
The epilepsy cohort consisted of 9061 patients. Mean age was 54.3 years (range, 15-97 years). A total of 4001 patients died during the follow-up period. The resultant SMR was 3.6, indicating that patients with seizure disorders were 3.6 times as likely to die during these 12 years than the general population.
Causes of death in the seizure patients can be categorized as: 1) causes of epilepsy; 2) complications of seizures; and 3) a few "surprises." Causes of epilepsy which also were causes of death were brain tumors, stroke, alcohol abuse, and dementia. Seizure complications were pneumonia, falls, drowning, and burns. Interestingly, motor vehicle accidents did not contribute to the increased SMR among the epilepsy cohort. Presumably, this reflects a reduced amount of driving among the seizure patients. More difficult to explain and surprising were the increased number of epilepsy-associated deaths due to the following: suicide, systemic cancer, poisoning, and digestive system diseases.
COMMENTARY
This important paper by Nilsson et al has several desirable and undesirable features. Some of these features were reviewed in the accompanying editorial by Nashef and Shorvon (Epilepsia 1997;38:1059-1061).
Nilsson et al’s report comprised a large number of patient years (52,520). Their sources of patients were more diverse than sources in previous institution- and referral-center-based studies. In 48% of cases, the causes of death were determined by autopsy. The epilepsy population was compared with a well-matched general population, all of whom were residents of Stockholm County over the same time period.
On the other hand, Nilsson et al identified their study cases from hospital discharge records. This may select for patients with seizures that are more severe or repetitive than the norm. It may also select for other co-morbid medical conditions. For example, an elderly patient with ischemic heart disease who has a single seizure may be more likely to be admitted to the hospital for "observation" than an otherwise well teen with new onset of juvenile myoclonic epilepsy. In fact, all cases of "epilepsy" are lumped together. Conditions such as idiopathic (genetic) primary generalized epilepsy and symptomatic localization-related epilepsy might well have quite different associated mortalities, but these were not studied separately. Finally, the authors’ case identification methodology also may account for the high mean age of their population (54.3 years) and the high mortality rate (44%) during the 12-year study.
Nilsson et al, in their large and well-controlled study, confirm and detail the previously reported increased mortality associated with chronic epilepsy. In another recent interesting paper, O’Donoghe and Sander looked at mortality over the last century in the Chalfont (residential) Centre for Epilepsy (CCE) in the United Kingdom (Acta Neurol Scand 1997;96:138-141).
At CCE, the medical officer’s annual reports from 1896 through 1965 documented all deaths. O’Donoghe and Sander calculated SMRs at CCE for five-year intervals, comparing death rates in the Chalfont epilepsy "colony" with age- and sex-matched general population death rates from the Registrar General’s Statistical Review of England and Wales. Overall, since 1896, the epilepsy patient SMR was 2.34 (not very dissimilar from that reported above by Nilsson et al). Remarkably, in all but three of the 14 five-year intervals between 1896 and 1965, the SMR for the Chalfont colony residents was in the 1.3-3.5 range. No trend toward decreasing mortality was seen in the more recent years.
Taken together, these papers reinforce the notion that chronic epilepsy is associated with increased mortality. However, explanations to completely account for this excess mortality remain elusive. Finally, this mortality seems to have persisted unchanged from 1896 through 1965, despite advances in the treatment of epilepsy over that time period. drl
Subscribe Now for Access
You have reached your article limit for the month. We hope you found our articles both enjoyable and insightful. For information on new subscriptions, product trials, alternative billing arrangements or group and site discounts please call 800-688-2421. We look forward to having you as a long-term member of the Relias Media community.