Late-onset epilepsy is linked to a substantial increased risk of subsequent dementia.

Results of a retrospective analysis show that patients who develop epilepsy at age 67 or older have a threefold increased risk of subsequent dementia vs their counterparts without epilepsy.

Dr Emily L. Johnson

“This is an exciting area, as we are finding that just as the risk of seizures is increased in neurodegenerative diseases, the risk of dementia is increased after late-onset epilepsy and seizures,” study investigator Emily L. Johnson, MD, assistant professor of neurology at Johns Hopkins School of Medicine in Baltimore, Maryland, told Medscape Medical News. “Several other cohort studies are finding similar results, including the Veterans’ Health Study and the Framingham Study,” she added.

The study was published online October 23 in Neurology.

Bidirectional Relationship?

Previous research has established that dementia is a risk factor for epilepsy, but recent studies also suggest an increased risk of incident dementia among patients with adult-onset epilepsy.

Several risk factors for late-onset epilepsy, including diabetes and hypertension, also are risk factors for dementia. However, the effect of late-onset epilepsy on dementia risk in patients with these comorbidities has not been clarified.

To investigate, the researchers examined data from the Atherosclerosis Risk in Communities (ARIC) study. Participants include Black and white men and women from four US communities.

Baseline visits in this longitudinal cohort study began between 1987 and 1989, and follow-up included seven additional visits and regular phone calls.

The investigators identified participants with late-onset epilepsy by searching for Medicare claims related to seizures or epilepsy filed between 1991 and 2015.

Those with two or more such claims and age of onset of 67 years or greater were considered to have late-onset epilepsy. Participants with pre-existing conditions such as brain tumors or multiple sclerosis were excluded.

ARIC participants who presented in person for Visits 2, 4, 5, and 6 underwent cognitive testing with the Delayed Word Recall Test, the Digit Symbol Substitution Test, and the Word Fluency Test.

Testing at Visits 5 and 6 also included other tests, such as the Mini-Mental State Examination, the Boston Naming test, and the Wechsler Memory Scale-III. Johnson and colleagues excluded data for Visit 7 from the analysis because dementia adjudication was not yet complete.

The researchers identified participants with dementia using data from Visits 5 and 6 and ascertained time of dementia onset through participant and informant interviews, phone calls, and hospital discharge data. Participants also were screened for mild cognitive impairment (MCI) at Visits 5 and 6.

Data were analyzed using a Cox proportional hazards model and multinomial logistic regression. In subsequent analyses, researchers adjusted the data for age, sex, race, smoking status, alcohol use, hypertension, diabetes, body mass index (BMI), APOE4 status, and prevalent stroke.

The researchers found that of 9033 study participants, 671 had late-onset epilepsy.

The late-onset epilepsy group was older at baseline (56.5 vs 55.1 years; P < .001) and more likely to have hypertension (38.9% vs 33.3%), diabetes (16.1% vs 9.6%), and two alleles of APOE4 genotype (3.9% vs 2.5%) vs those without the disorder.

In all, 1687 participants developed dementia during follow-up. The rate of incident dementia was 41.6% in participants with late-onset epilepsy and 16.8% in participants without late-onset epilepsy. The adjusted hazard ratio of subsequent dementia in participants with late-onset epilepsy vs those without the disorder was 3.05 (95% CI, 2.65 – 3.51).

The median time to dementia ascertainment after late-onset epilepsy was 3.66 years.

Counterintuitive Finding

Interestingly, the relationship between late-onset epilepsy and subsequent dementia was stronger in patients without stroke.

The investigators offered a possible explanation for this counterintuitive finding.

“We observed an interaction between [late-onset epilepsy] and stroke, with a lower (but still substantial) association between [late-onset epilepsy] and dementia in those with a history of stroke. This may be due to the known strong association between stroke and dementia, which may wash out the contributions of [late-onset epilepsy] to cognitive impairment,” the researchers write.

“There may also be under-capturing of dementia diagnoses among participants with stroke in the ascertainment from [Centers for Medicare & Medicaid Services] codes, as physicians may be reluctant to make a separate code for ‘dementia’ in those with cognitive impairment after stroke,” they add.

When the researchers restricted the analysis only to participants who attended Visits 5 and 6 and had late-onset epilepsy ascertainment available, they found that the relative risk ratio (RRR) for dementia at Visit 6 was 2.90 (95% CI, 1.22 – 6.92; = .009). The RRR for MCI was 0.97 (95% CI, 0.39 – 2.38; = .803). The greater functional impairment in patients with late-onset epilepsy may explain the lack of a relationship between late-onset epilepsy and MCI.

“It will be important for neurologists to be aware of the possibility of cognitive impairment following late-onset epilepsy and to check in with patients and family members to see if there are concerns,” said Johnson.

“We should also be talking about the importance of lowering other risk factors for dementia by making sure cardiovascular risk factors are controlled and encouraging physical and cognitive activity,” she added.

The results require confirmation in a clinical population, the investigators note. In addition, future research is necessary to clarify whether seizures directly increase the risk of dementia or whether shared neuropathology between epilepsy and dementia explains the risk.

“In the near future, I plan to enroll participants with late-onset epilepsy in an observational study to better understand factors that may contribute to cognitive change. Collaborations will be key as we seek to further understand what causes these changes and what could be done to prevent them,” Johnson added.

Strengths, Weaknesses

In an accompanying editorial, W. Allen Hauser, MD, professor emeritus of neurology and epidemiology at Columbia University in New York, and colleagues note the findings support a bidirectional relationship between dementia and epilepsy adding that accumulation of amyloid beta peptide is a plausible underlying pathophysiology that may explain this relationship.

Future research should clarify the effect of factors such as seizure type, seizure frequency, and age of onset on the risk of dementia among patients with epilepsy, the editorialists write. Such investigations could help elucidate the underlying mechanisms of these conditions and help to improve treatment, they add.

Commenting on the findings for Medscape Medical News, Ilo Leppik, MD, professor of neurology and pharmacy at the University of Minnesota in Minneapolis described the research as “a very well-done study by qualified researchers in the field.

“For the last century, medicine has unfortunately become compartmentalized by specialty and then subspecialty. The brain and disorders of the brain do not recognize these silos…. It is not a stretch of the known science to begin to understand that epilepsy and dementia have common anatomical and physiological underpinnings.”

The long period of prospectively gathering data and the measurement of cognitive function through various modalities are among the study’s great strengths, said Leppik.

However, the study’s weakness is its reliance on Medicare claims data, which mainly would reflect convulsive seizures.

“What is missing is how many persons had subtle focal-unaware seizures that may not be identified unless a careful history is taken,” said Leppik. “Thus, this study likely underestimates the frequency of epilepsy.”

Neurologists who evaluate a person with early dementia should be on the lookout for a history of subtle seizures, said Leppik. Animal studies suggest treatment with levetiracetam or brivaracetam may slow the course of dementia, and a clinical study in participants with early dementia is under way.

“Treatment with an antiseizure drug may prove to be beneficial, especially if evidence for the presence of subtle epilepsy can be found,” Leppik added.

Greater collaboration between epileptologists and dementia specialists and larger studies of antiseizure drugs are necessary, he noted.

“These studies can incorporate sophisticated structural and biochemical [analyses] to better identify the relationships between brain mechanisms that likely underlie both seizures and dementia. The ultimate promise is that early treatment of seizures may alter the course of dementia,” Leppik said.

The study by Johnson and colleagues was supported by a contract from the National Institute on Aging; ARIC from the National Heart, Lung, and Blood Institute; the National Institutes of Health; and the Department of Health & Human Services. The authors and Leppik have disclosed no relevant financial relationships.

Neurology. Published online October 23, 2020. 

SOURCE: Medscape Neurology News by E. Greb