Article In Brief
Investigators sought to examine the complexity and range of patient complaints with regard to the most common epilepsy drugs and identify which, if any, factors stood out as predictors of cognitive function.
Although a number of other variables can contribute to cognitive dysfunction in patients taking standard antiepileptic drugs (AEDs), an observational, cross-sectional study has found none to be independently predictive of such problems.
“We found no antiepileptic drug to be independently associated with cognitive dysfunction, although seizure frequency and depression were linked to objective and subjective cognitive dysfunction, respectively,” said lead author Emma Foster, MBBS, an epilepsy fellow and neuroscience researcher at Monash University and Alfred Hospital, in Melbourne, Australia. “Our findings suggest that optimizing therapy to prevent seizures, while having many benefits for the patient, may not result in significant impact on cognitive function.”
However, epilepsy experts contacted by Neurology Today were not convinced by the findings and said further research in better-designed studies is necessary.
Dr. Foster and her colleagues evaluated 331 epilepsy patients at an EEG monitoring unit. Nearly half had received AED polypharmacy, about one quarter had received monotherapy, and the remainder were untreated. Objective cognitive function was scored using standardized testing, while subjective cognitive function and quality of life were determined via patient questionnaires, as were signs of anxiety and depression.
The most frequent AEDs, in descending order, were levetiracetam, valproate, carbamazepine, lamotrigine, and topiramate.
Advanced age, seizure frequency, and concomitant epilepsy and psychogenic non-epileptic seizure (PNES) were predictors of objective cognitive function, while depression, anxiety, and seizure frequency were predictors of subjective cognitive function. Impaired cognitive function above or beyond these and other clinical variables was not associated with AEDs.
The study was published in the XX issue of Neurology.
“Our study sample was heterogeneous and represented a broad range of diagnoses, treatment regimens, and comorbidities,” Dr. Foster told Neurology Today. “Such heterogeneity is a significant strength because it has high external validity and can be translated into clinical practice. This real life ‘messy’ setting is the reality of the tertiary care practice.”
While the study assessed AED effects within the context of other factors known or suspected to impact cognition, it was not a randomized controlled trial, and any inference regarding a causal relationship between AEDs and cognition is not yet possible, she emphasized.
“It is possible that complex, second-order relationships between variables have obscured the true relationships between specific AEDs and cognition. Nonetheless, our observation suggests that, within the limitations of the cross-sectional study design, specific AEDs do not have independent relationships with cognition above and beyond other key clinical variables.”
Topiramate, in particular, has been associated with cognitive complaints, and well-designed studies have associated its use with declining verbal fluency, attention and concentration, processing speed, language skills, perception, and working memory, Dr. Foster noted. However, she pointed to important differences between earlier research and the new study's design and population.
For example, one randomized double-blind study concluded that topiramate-related neuropsychological impairment emerged in a dose-dependent fashion, but the study cohort consisted of healthy subjects rather than those with epilepsy, she explained, adding that other studies have had smaller sample sizes and less rigorous measures of cognitive dysfunction.
“In direct contrast to the existing literature, our study did not find ... topiramate predictive of objective or subjective cognitive function above and beyond other relevant clinical factors. Importantly, our analyses confirmed that statistical power was not a likely explanation for the findings of no association between specific AED and objective or subjective cognitive function,” Dr. Foster said.
Dr. Foster said the findings of a large post-marketing study support this. “It found, despite cognitive complaints being the most common reason for topiramate discontinuation, most patients still chose to continue topiramate therapy beyond six months, stating it conferred better seizure control and overall improvement than pre-topiramate state.”
Moreover, several factors have been associated with reduced topiramate related-adverse cognitive effects, she said, including lower dosage, and patient factors, such as higher educational attainment and later onset of epilepsy.
“These are important points to bear in mind when considering AED choice and dose,” she told Neurology Today.
“Our findings should perhaps reassure patients and clinicians that the real-world effect of topiramate may not be as substantial as the literature suggests, and it remains a useful AED.”
Claire S. Jacobs, MD, PhD, an instructor of neurology at Massachusetts General Hospital, told Neurology Today she was surprised by the findings because the weight of evidence indicates an association between AEDs and cognitive dysfunction.
“These results directly contradict numerous previous studies that support a link between several AEDs, such as topiramate, and cognitive issues,” she said. “While I am intrigued by this study, I believe the weight of the current data supports an influence of AEDs on cognitive function.”
However, she added that her experience is consistent with the reported findings between a decline in cognitive function and increasing age, higher seizure frequency, PNES, and poorly controlled depression and anxiety.
“My clinical experience is that cognitive side effects are quite common, dose dependent, and are often more prominent with increasing age and polypharmacy. I generally avoid prescribing AEDs that are sedating or have been linked to cognitive dysfunction. This article serves as a reminder of the importance of focusing not only on seizure control, but also on adequate detection and treatment of depression and anxiety in people with epilepsy.”
Dr. Jacobs said cognitive issues are among the most frequent concerns that she hears from her patients, particularly following a breakthrough seizure.
“I typically avoid cognitively ‘dirty’ AEDs, especially in older individuals or those with cognitively demanding professions or lifestyles. I also discuss modifiable lifestyle factors, such as sleep, exercise, and substance use, that can affect cognitive function.”
Kimford J. Meador, MD, FAAN, professor of neurology at Stanford University School of Medicine in Palo Alto, CA, also said that he remains unpersuaded.
“Although this is interesting data from a strong investigator group, the study design is not appropriate to assess cognitive effects,” he told Neurology Today.
He noted, for example, that patients were not randomized to their specific AED, which, he said, creates selection bias. Also, because the cognitive analysis was performed at only one time point, test-retest effects could not be determined. Moreover, he said cross-sectional analysis may confound the effects of AEDs with other major factors that can contribute to cognitive performance, including the underlying cause of epilepsy and baseline cognitive abilities.
AED cognitive effects can be dose-dependent, he continued, but the study did not assess AED dosages and blood levels. These design issues, as well as others, likely explain why the findings are in conflict with multiple other studies, he told Neurology Today.
Responding to the comments, Dr. Foster told Neurology Today that the aim of the study was not to determine whether or not a causal relation exists between AEDs and cognitive issues, but to examine the complexity and range of patient complaints with regard to the most common epilepsy drugs and identify which, if any, factors stood out as predictors of cognitive function.
“As Dr. Meador points out, this study design is not appropriate to demonstrate causation, and was not intended to do so,” she said. “Instead, our study is a snapshot of hundreds of patients with a wide variety of factors that may impact cognition, such as age, epilepsy type and severity, psychiatric comorbidities, and presence and type of AEDs.”
She said the researchers chose the study design to include a broad range of patients representing the real-life clinical practice in a tertiary epilepsy care setting.
“Any factors that are consistently associated with cognitive impairment in this heterogeneous cohort are worthy of clinical consideration,” Dr. Foster said. “The factors we identified as predictors of subjective cognitive complaints were mood disorders and seizure frequency. The factors associated with objective cognitive impairment were increasing age, concomitant epilepsy and PNES diagnoses, and seizure frequency.”
These factors have also been identified as predictors of cognitive impairment in other studies, she noted. “For example, anxiety and depression have been associated with subjective cognitive impairment, as Professor Meador observed. This reinforces the need for clinicians to diagnose and adequately treat mood disorders and seizure frequency in patients with cognitive complaints, as they are modifiable factors.”
Although, AEDs were not independent predictors of impaired cognition in the study, she said the investigators acknowledge that AED cognitive effects may be significant for a subset of patients.
“Our findings are in opposition to much of the existing literature, and we agree with Dr. Meador that this may be due to our study design.”