Timing of Epilepsy Onset Influences Type of Cognitive Impairment and Memory Deficits

Patients with early- versus late-onset epilepsy exhibited distinct differences in the origin of memory deficits, supporting the practice of subtyping patients by clinical characteristics to identify individualized predictors of cognitive dysfunction, according to the results of a new study (Rayner G, et al. Neurology. 2016;87:1642-1649).

Neurobiologic factors, such as younger age at onset, more frequent seizures, and reduced working memory, had a strong association with autobiographic memory dysfunction in patients with early-onset epilepsy. Conversely, memory impairment associated with late-onset epilepsy arose from psychological maladjustment, including depression, as well as magnetic resonance imaging (MRI)-detected brain lesions, reported Genevieve Rayner, MPsych, PhD, Honorary Fellow and Sessional Teaching, Melbourne School of Psychological Sciences, University of Melbourne, Australia, and colleagues.

“The results of this study support a life-span approach to understanding the cognitive impairments common to neurology, taking into account the deleterious effects of seizures on the developing brain, as well as psychological processes associated with adjusting to a chronic illness,” Dr Rayner and colleagues noted. “Different predictors of AMN [autobiographic memory network] dysfunction in different epilepsy syndromes support the use of neurocognitive phenotyping as a valid method to generate more individualized neurologic research,” they concluded.

“Crucially, the close interrelation of cognition and behavior in epilepsy suggests that these functions may share neural substrates, and questions the utility of studying them as independent epiphenomena of epilepsy,” the researchers added.

The authors of an accompanying editorial described the observed association between the timing of epilepsy onset and the factors implicated in memory impairment “a remarkable finding” that could inform future investigation and clinical practice (Barr WB, Jones JE. Neurology. 2016;87:1634-1635).

“If the field continues to develop an understanding of the neurologic, cognitive, and neuropsychiatric phenotypes of epilepsy, we may be able to identify the neural substrates that are involved and provide a better understanding of how epilepsy affects function across the life span,” concluded William B. Barr, PhD, Director of Neuropsychology, NYU Langone Medical Center, New York, NY, and Jana E. Jones, PhD, Associate Professor of Neuropsychology, University of Wisconsin School of Medicine and Public Health, Madison.

Memory Deficits in Early-Onset versus Late-Onset Epilepsy

Cognitive impairment, particularly autobiographic deficits, occurs frequently in patients with epilepsy. Patients with focal epilepsy perform poorly on tasks that involve recalling personal information across their life­span compared with healthy individuals, the authors noted. Marked memory deficits often occur despite only mild decrements on standard tests of verbal and visual memory.

The identification of neurocognitive phenotypes for patients with epilepsy could lead to the increased individualization of clinical management and treatment. Emerging evidence has suggested potentially different mechanisms for autobiographic memory deficits during childhood-onset versus adult-onset epilepsy, the researchers noted.

The goal of the study was to examine whether the origins of memory deficits in patients with epilepsy differed according to the timing of disease onset.

“We hypothesized that impairments in individuals with seizures emerging in a critical neurodevelopmental period of childhood/adolescence will be linked to epilepsy-related factors,” Dr Rayner and colleagues noted. “In contrast, reduced autobiographic recollection in recent-onset epilepsy will be linked to nonepilepsy factors such as depression.”

The study involved 92 adults with focal epilepsy and 74 individuals with no neurologic or psychiatric history who were primarily recruited from the patients’ families, which facilitated sociodemographic matching. Participants in both groups completed the same neuropsychologic evaluation, which included assessments of autobiographic memory, broader memory function (such as the Wechsler Memory Scale), and mood.

Dr Rayner and colleagues examined correlates of autobiographic memory impairment within the neurodevelopmental context of early- versus late-onset epilepsy. The mean age was 37.5 years in patients with early-onset epilepsy versus mean age of 44.7 years (P = .006) in patients with late-onset epilepsy. In addition, patients with early-onset epilepsy had a longer duration of epilepsy compared with patients with late-onset epilepsy (26.1 years vs 12.6 years, respectively; P <.001).

Among the subgroup of 47 patients with early-onset epilepsy, memory deficits were associated most closely with monthly seizure frequency (P = .012), which accounted for 11% of the variance in predicting semantic memory impairment in a multiple regression analysis (P = .025). Poor recall of salient life events was also associated with younger age at epilepsy onset (P = .032) and lower Symbol Span score (P = .046). Combining Symbol Span score and the age at epilepsy onset resulted in a model that accounted for 17% of the variance (P = .047).

Among the 45 patients with late-onset epilepsy, poor semantic memory was associated with higher depression scores (P = .007). Cognitive correlates included immediate recall (P = .006), delayed recall (P = .011), and Symbol Span score (P = .034). The presence of MRI-detected lesions also correlated with poor recall of semantic facts (P = .047), as did treatment with multiple medical therapies as opposed to monotherapy (P = .033). Depression scores alone accounted for 15.4% (P = .015) of the variance, increasing to 29.4% (P = .002) of the variance with the addition of lesion status.

Poor recollection of autobiographic memories was significantly associated with depression (P = .003). Cognitive correlates included 2 scales of verbal recall (P = .031 and P = .024, respectively), immediate recall (P = .041), delayed recall (P = .008), and Symbol Span score (P = .038).

Depression scores accounted for 19.5% of the variance in patients with late-onset epilepsy (P = .005).