Epilepsy in Alzheimer’s Can Be Early and Subtle

Studies have established ties between epilepsy and Alzheimer’s disease (AD), but how the two relate in time and what exactly is the nature of seizures in AD patients remain poorly understood. A report published online July 8 in JAMA Neurology addresses these questions by examining patients with epileptic brain activity and either amnestic mild cognitive impairment (aMCI) or AD. Researchers led by Lennart Mucke and Keith Vossel of the Gladstone Institute of Neurological Disease, San Francisco, California, found that in these patients, as was previously found in mouse models, epileptic activity often manifests as non-convulsive seizures. They also report that these types of seizure are associated with earlier cognitive decline. Though preliminary, the results could steer clinicians toward using certain anticonvulsant medications for the treatment of AD in the future.

“Seizures represent an aspect of Alzheimer’s disease that may be treatable,” Vossel told Alzforum. “If nothing else, it could improve symptoms, but in the best-case scenario, we might actually target some mechanisms at play early in the disease process by suppressing aberrant network activity.”

Many studies have characterized epileptic seizures as an endpoint in the late stages of Alzheimer’s (see review by Mendez et al., 2003 and Romanelli et al., 1990). These studies typically described convulsive seizures in people already diagnosed with AD. First author Keith Vossel and colleagues wanted to know if epileptic activity occurred earlier in the disease, how it manifested, what regions of the brain were most affected, and whether it might be helpful in diagnosing AD.

The researchers collaborated with Bruce Miller of the Memory and Aging Center at the University of California, San Francisco, to undertake a retrospective study in which they searched the center’s database for patients diagnosed with amnestic mild cognitive impairment (aMCI) or AD between 2007 and 2012. The scientists narrowed their selection to those with a diagnosis of epilepsy or evidence of epileptic brain activity. They excluded anyone whose seizures had started at a young age or who had other epilepsy risk factors such as stroke or a tumor. Out of 233 people with aMCI, 12 had developed epilepsy after the age of 30. Of the 1,024 diagnosed with AD, 35 had epilepsy and seven more showed signs of epileptiform activity on electroencephalographs (EEG). Mucke emphasized that these numbers likely underestimate the real incidence of epileptiform activity in aMCI and AD, because most patients included in this study did not undergo extensive testing by the most sensitive methods available.

On average, cognition in people with epilepsy declined about five to seven years earlier than in people with normal brain activity. Decline began before age 65 in half of those with epilepsy, but in only about a quarter of those without. Seizures often began around the same time cognition began to deteriorate, when neuropsychological deficits were still mild. Records showed about half of the epilepsy patients had only non-convulsive seizures, exhibiting sensory distortions, speech problems, emotional abnormalities, or feelings of déjà vu. Electroencephalography (EEG) had picked up epileptiform activity in 62 percent of patients with seizures and 6 percent of people who underwent EEG for other reasons.

Epileptic foci were located most commonly in the temporal cortex, followed by the frontotemporal, frontal, and central cortices. In general, long-term video or serial EEGs detected epileptiform activity better than did routine EEG.

That epileptic activity commonly affected the temporal lobes makes it likely that it contributes to memory impairment in these patients, the authors claim. However, its contribution to the overall cognitive decline in AD remains to be determined. “We do not know the extent to which seizures are causing cognitive loss in this particular population,” said Michela Gallagher, John’s Hopkins University, Baltimore, Maryland, who was not involved in the study. Gallagher previously found that reducing hippocampal hyperactivity with a low dose of the anticonvulsant drug levetiracetam improved hippocampal function and memory in people with MCI (see ARF related news), suggesting that such medications could improve cognition.

Studies in transgenic mouse models of AD support that idea. Mucke previously reported that dampening neuronal hyperactivity improves cognition in J20 mice (see ARF related news). That suggests aberrant network excitability could lead to cognitive dysfunction, Mucke told Alzforum. What leads to this hyperactivity? Previous work indicates that in mouse models elevated soluble Aβ appears to be a factor (see ARF related news).

If anti-epileptic treatments might improve cognition, then which drug might work best? Vossel and colleagues found that in the UCSF patient population, some anticonvulsants performed better than others Lamotrigine and levetiracetam, when used at low doses, either partially or fully relieved seizures in about 94 percent of the epilepsy patients. Phenytoin and valproate, which failed in a phase 3 AD trial (see ARF related news) proved less tolerable, and provided less relief. “Together with our previous data from AD animal models, these observations suggest that long-term phenytoin treatment is probably best avoided in this patient population,” Mucke told Alzforum (see Verret et al. 2012 and Sanchez et al. 2012).

The authors acknowledged that this was a small study, relying on data from a single memory center that specializes in early onset dementia, and on notes that lacked standardized methods. Vossel is leading an ongoing prospective study enrolling healthy controls, people with MCI, and AD patients. Using 36-hour video-EEG monitoring and magnetoencephalography (MEG), the scientists hope to identify a subpopulation who might benefit from epilepsy-related diagnostics and medication. So far, in cohorts enriched with early onset AD cases, epileptiform abnormalities are seen in nearly half of those tested. In the future, the scientists plan to spin off a pilot study in which some people with epileptic activity receive levetiracetam.

“This is a much-needed exploration into the association of degenerative disease with seizures,” wrote Gregory Jicha, University of Kentucky Medical Center, Lexington, to Alzforum in an email. However, he cautioned against drawing conclusions about antiepileptic treatments based on this small retrospective study. He also pointed out that no data have yet shown that improved seizure control helps cognitive or behavioral symptoms in dementia.—Gwyneth Dickey Zakaib

Comments

  1. These authors represent the major, if not the only, group of researchers devoted to analysis of Aβ’s effects on inhibitory/excitatory balance, neuronal sprouting, characteristics of epileptiform activity in the temporal lobe subregions, neuronal network alteration, and how these correlate with cognitive decline both in animal models and in humans. This article is, to my knowledge, the only one to compare epileptic and non-epileptic MCI subjects with respect to their cognitive decline and treatment outcome. A potentially intriguing finding is that amnestic MCI subjects lack subclinical epileptiform EEG activity compared to controls.

    The authors are developing an innovative and interesting approach to researching Alzheimer's disease causes. There are, however, a few aspects of the study design that complicate interpretation. First, there is no reliable measure of cognitive outcome to fully demonstrate any impact of antiepileptic treatment on the progression of cognitive decline. This would be most informative. Second, the lack of genetic data on the ApoE polymorphisms (only 11 of the 60 patients enrolled) is a pity since Apoe E4/E4 is a risk factor for both epilepsy and cognitive decline and might be responsible for the fact that aMCI and AD patients with epilepsy have an earlier onset and a faster progression. Finally, the authors present no clear clinical evidence for diagnostic certainty (PIB imaging only for 9 patients, spinal fluid markers only for 3, and no post-mortem analysis).

    View all comments by Paolo Maria Rossini

  2. The evidence is clear that seizures plus dementia are worse than dementia alone. Poorly controlled seizures are well known to induce neuronal loss and inflammation in the hippocampus, affecting memory circuits (medial temporal lobe sclerosis or hippocampal sclerosis). The present study moves these findings forward into earlier stages of disease, such as mild cognitive impairment (MCI). Raising suspicion of subclinical or partial seizure rather than focusing on pure convulsive seizures is important clinically. Since only 7 percent of MCI (17/233) and 6 percent of AD patients (63/1024) were found to have seizure activity in this study, routine EEG for the purpose of detecting seizures without clinical suspicion is not warranted given the low pre-test probability. In subjects reporting paroxysmal events (transient cognitive/language/behavior deficits) that resolve, consideration of EEG is important clinically, with the caveat that neither the present paper nor others in the literature have ever shown improvement in cognitive/behavioral/psychiatric symptoms of dementia with improved seizure control or the use of anti-epileptic drugs.

    EEG may be more globally useful to detect degenerative disease by focusing on slowing and other non-epileptiform characteristics, although this remains to be proven. Our work suggests that algorithms can be created using EEG patterns that can detect early cognitive impairment. These effectively act as biomarkers of disrupted neuronal connectivity and function, before overt structural biomarkers such as MRI can detect such changes.

    I would be cautious about making recommendations for anti-epileptic drugs based on retrospective analysis of their use in such a small cohort. While the authors discuss this caveat, they still make a strong recommendation for levitiracetam or lamotrigine. Only prospective clinical trials will tell if these drugs may be beneficial . Trials of drugs in subjects with dementia/MCI and seizures is warranted given the prevalence of such cases in the U.S. population. Given more than five million people have AD, and an estimated seven million have MCI, about 720,000 people would potentially benefit from a successful therapy in the US alone.

    The earlier onset of dementia in cases with seizures is expected from what we know of mixed dementia states. Data shows that pathologies are additive. We recently examined brain bank data on subjects with MCI or dementia who had had seizures, and, after controlling for the degree of cognitive impairment, we found that those with seizures had significantly lower pathology than those without seizures, suggesting that the seizure disorder added to the degenerative AD pathology to exacerbate the cognitive decline. (We have presented this, but it is not yet published.) Thus, the earlier age of onset for cognitive symptoms if patients have concomitant seizure disorder. It makes sense to aggressively identify and treat concomitant seizures in dementia, but again proven benefit of such an approach is lacking despite strong supportive data and medical rationale.

    Overall this is a much-needed exploration into the association of degenerative disease, even in a prodromal phase, with seizures. If identifying such patients and treating them could delay MCI by seven years and AD by five years, in just 6 percent of patient, this would be a major victory for hundreds of thousands in the U.S. and millions worldwide.

    View all comments by Gregory Jicha

Make a Comment

To make a comment you must login or register.

References

News Citations

  1. Epilepsy Drug Calms the Hippocampus, Aids Memory
  2. Anticonvulsants Reverse AD-like Symptoms in Transgenic Mice
  3. Soluble Aβ Takes Blame for Hyperactive Neurons in Mouse Brain
  4. Chicago: AD and Epilepsy—Lessons from the Clinic, Animals

Paper Citations

  1. Mendez M, Lim G. Seizures in elderly patients with dementia: epidemiology and management. Drugs Aging. 2003;20(11):791-803. PubMed.
  2. Romanelli MF, Morris JC, Ashkin K, Coben LA. Advanced Alzheimer's disease is a risk factor for late-onset seizures. Arch Neurol. 1990 Aug;47(8):847-50. PubMed.
  3. Verret L, Mann EO, Hang GB, Barth AM, Cobos I, Ho K, Devidze N, Masliah E, Kreitzer AC, Mody I, Mucke L, Palop JJ. Inhibitory interneuron deficit links altered network activity and cognitive dysfunction in Alzheimer model. Cell. 2012 Apr 27;149(3):708-21. PubMed.
  4. Sanchez PE, Zhu L, Verret L, Vossel KA, Orr AG, Cirrito JR, Devidze N, Ho K, Yu GQ, Palop JJ, Mucke L. Levetiracetam suppresses neuronal network dysfunction and reverses synaptic and cognitive deficits in an Alzheimer's disease model. Proc Natl Acad Sci U S A. 2012 Oct 16;109(42):E2895-903. PubMed.

Other Citations

  1. J20 mice

Further Reading

No Available Further Reading

Primary Papers

  1. Vossel KA, Beagle AJ, Rabinovici GD, Shu H, Lee SE, Naasan G, Hegde M, Cornes SB, Henry ML, Nelson AB, Seeley WW, Geschwind MD, Gorno-Tempini ML, Shih T, Kirsch HE, Garcia PA, Miller BL, Mucke L. Seizures and epileptiform activity in the early stages of Alzheimer disease. JAMA Neurol. 2013 Sep 1;70(9):1158-66. PubMed.

Annotate

To make an annotation you must Login or Register.