Is Late-Onset Epilepsy a Vascular Warning Sign?

Smoking, hypertension, diabetes, chronic kidney disease, APOE epsilon 4, and prior stroke were each independently associated with higher risk for late-onset epilepsy (LOE) in a new pooled analysis of older adults, a pattern that suggests the condition may be a marker of underlying vascular disease in the brain.The results add to an ongoing debate over whether a diagnosis of LOE should prompt screening for undiagnosed vascular risk factors or more aggressive management of known ones.Identifying vascular risk factors such as chronic kidney disease — a link not explored before — is important because LOE is expected to become more common as the population ages. Researchers found that LOE incidence among adults aged 65 years or older was two- to threefold higher than that of Parkinson’s disease and approaching that of stroke.Hyunmi Choi, MD, MS“We have a tendency to look at those individuals who develop LOE as people who just happen to develop seizures later in life, and our focus is primarily on seizure control,” lead investigator Hyunmi Choi, MD, MS, an epileptologist in the Department of Neurology at Columbia University Irving Medical Center in New York City, told Medscape Medical News.“I think the shift in how we approach this population is to look at them and consider the underlying vascular burden that they might have,” she added.The findings were published online on October 19 in JAMA Neurology.LOE Underrecognized in Older AdultsLOE can be easy to miss. The condition has historically been underrecognized, Choi said, as epilepsy research focused on younger patients with drug-resistant temporal lobe epilepsy.“Some of these people, I think, can present with just transient alteration in awareness,” Choi added, saying clinicians may attribute such episodes to a transient ischemic attack or progressive dementia.Earlier single-cohort studies have linked hypertension, diabetes, hypercholesterolemia, and smoking to LOE, but those cohorts had limited racial and ethnic diversity, the authors noted.To address those gaps, researchers combined individual-level data from six US population-based cohorts whose primary purposes included cardiovascular and cerebrovascular health. Follow-up data spanned from 1991 to 2024.The final analysis included 42,257 adults (mean age, 71.4 years; 56.0% women) with no history of epilepsy. About 26% of the participants were non-Hispanic Black and 10% were Hispanic.In five cohorts, epilepsy was identified through Medicare claims using validated definitions that combined diagnostic codes with antiseizure medication use. The sixth cohort used medical record review with adjudication by a study neurologist.Risk factors were assessed at entry into each cohort through self-reports, examinations, laboratory tests, and medication records, and the findings were adjusted for cohort, demographics, and all other covariates.Vascular Risk FactorsIncidence rose with age and was the highest among non-Hispanic Black participants, at 6.4 cases per 1000 person-years. Vascular risk factors did not explain that disparity, the authors wrote, and associations were similar across age, sex, and race and ethnicity subgroups.History of stroke was associated with nearly double the risk for LOE (adjusted hazard ratio [aHR], 1.99; 95% CI, 1.63-2.43). Diabetes (aHR, 1.42; 95% CI, 1.25-1.62), current smoking (aHR, 1.39; 95% CI, 1.18-1.64), chronic kidney disease (aHR, 1.29; 95% CI, 1.13-1.48), and hypertension (aHR, 1.17; 95% CI, 1.04-1.31) were also associated with higher risks for LOE. Hypercholesterolemia, BMI, and history of myocardial infarction were not associated with LOE after full adjustment.Carrying one and two apolipoprotein E epsilon 4 alleles was associated with 22% and 84% higher risks for LOE, respectively. The variant is strongly associated with Alzheimer’s disease (AD), and the authors proposed that it may contribute to LOE through AD-related changes, vascular brain injury, or both.Choi described chronic kidney disease as a likely marker of systemic vascular disease but not a direct cause of LOE.In models of disease severity, each 10-mg/dL increase in fasting glucose was associated with a 3% higher risk for LOE. In models that also included medication use, glucose- and blood pressure-lowering medications were associated with 75% and 18% higher risks for LOE, respectively.The investigators acknowledged that claims-based diagnoses were not clinically confirmed, stroke history was self-reported, and risk factors were not updated after cohort entry, which for many participants came years before epilepsy follow-up began.Because risk factors were measured only at cohort entry, some participants counted as free of hypertension or diabetes developed those conditions later without being reclassified. Choi said this could mean the true associations are stronger than those observed.Experts See a Clearer SignalThe pooled data clarify the link between vascular risk factors and LOE in ways previous smaller cohorts couldn’t, said Rani A. Sarkis, MD, MSc, an epileptologist and behavioral neurologist at Mass General Brigham and associate professor of neurology at Harvard Medical School in Boston who was not part of the research.Rani A. Sarkis, MD, MSc“Factors tied to vascular health are putting you at risk of epilepsy later in life,” he told Medscape Medical News.Seizures in older adults often don’t always resemble the generalized tonic-clonic events clinicians expect, Sarkis said. Patients may have transient epileptic amnesia, “where the only symptom is they lose some of their memories,” confusion that lasts hours to days after a seizure, or episodes that look like panic attacks or hot flashes, he said.“A lot of them, they’re having these symptoms a year or two before their first tonic-clonic seizure brings them to medical attention,” he added.Because the study could count only diagnosed cases of LOE, Sarkis said its incidence rate “is probably an underestimate.” He added, “Some of these patients actually live alone, so they could have seizures without anyone knowing.”He also noted that without brain imaging, the study could not separate poststroke epilepsy from unexplained cases. “They don’t have the scans to really tell us who has a stroke, even if silent, and who doesn’t,” he said.Johan Zelano, MD, PhDJohan Zelano, MD, PhD, senior physician in neurology at Sahlgrenska University Hospital in Gothenburg, Sweden, whose group has mainly studied late-onset seizures as a predictor of stroke, called the study’s methodology “very rigorous…. Before the big-data era, it was really hard to figure out: Is it a coincidence or is it really an association?” Zelano was also not involved in the study.Whether lowering vascular risk prevents LOE remains untested, said Zelano. “Does a healthier vascular lifestyle, primary risk factor modification, alter the risk of late-onset epilepsy? That is unanswered, as far as I know,” he said.Sarkis said he has already changed how he approaches these visits. When he sees a patient with LOE in the clinic, Sarkis said, seizure control is only part of the appointment.“Don’t just focus on giving them a medication and controlling their seizures but also think about their risk factors,” he said. “This is a risk stratification interaction. ‘I’m going to meet you now and we’re going to talk about seizures, but I also want to figure out what your vascular risks are and try to address them.’”The Epilepsy-Cog study was supported by the National Institutes of Health. Choi reported receiving grants from the National Institute on Aging during this study. Disclosures for the other authors are listed in the original article. Sarkis reported previously receiving consulting fees from UCB. Zelano reported consulting for companies in the epilepsy field.