Prevalence and Factors Influencing Visual Memory Dysfunction among Epilepsy Patients—A Single-Center Study
Correspondence Address: Source of Support: None, Conflict of Interest: None DOI: 10.4103/0028-3886.289011
Source of Support: None, Conflict of Interest: None
Keywords: Epilepsy, predictors, prevalence, visual memory
Epilepsy is defined as at least two unprovoked seizures occurring more than 24 h apart or one unprovoked seizure and a probability of further seizures similar to the general recurrence risk (at least 60%) after two unprovoked seizures, occurring over the next 10 years or a diagnosis of an epilepsy syndrome. Quality of life among epilepsy patients is affected due to the physical injury sustained from seizures or cognitive dysfunction associated with poor school performance and employment.,
Cognition consists of remembering new information, attention focus, and problem-solving. Cognitive impairment may occur due to the disease itself or treatment received. In patients with epilepsy, memory appears to be the most vulnerable cognitive dysfunction. Memory is classified into short-term and long-term memory. Short-term memory is a transient trace of information that is temporarily stored and requires consolidation by the medial temporal lobe to be converted into long-term memory. Temporal lobe epilepsy (TLE) is more likely to cause long-term memory impairment due to the anatomical location of the hippocampus. Visual memory dysfunction is defined by impairment of any form of memory, in which the stored information is acquired initially by visual stimuli or unable to maintain the perceptual properties of the viewed stimuli.
The exact contributing factors that have an impact on visual memory dysfunction are still inconclusive. Apart from the anatomical aspect, longer duration of uncontrolled seizures, early age of onset, increased frequency of seizures, the higher dosage of antiepileptic drugs (AEDs), and abnormal epileptiform discharges on the electroencephalogram (EEG) contribute to the more adverse cognitive outcome. Detection of memory impairment among epilepsy patients is important as this dysfunction may be improved by cognitive rehabilitation.
Our present study was conducted to determine the prevalence of visual memory dysfunction among patients with epilepsy at a tertiary center. We also identified specific risk factors for visual memory dysfunction (such as types of epilepsy, duration of epilepsy, age of onset, the total number of AEDs, abnormal EEG upon diagnosis, and abnormal brain imaging).
This was a cross-sectional study involving patients with epilepsy from 16 to 69 years of age from the Neurology Clinic, Universiti Kebangsaan Malaysia Medical Centre from June 2018 until December 2018.
This study was conducted after obtaining approval from the research ethics committee (study code: FF-2017-439). All study participants were provided a copy of the patient information sheet and written consent prior to starting the study test. Consent was obtained from parents for patients age less than 18 years. A total of 496 patients with epilepsy were screened and 250 patients recruited based on the inclusion and exclusion criteria. The diagnosis of epilepsy was based on clinical history supported by EEG with brain-imaging evidence. All patients were on AEDs at the time of their evaluation. Brain imaging such as computed tomography (CT) or magnetic resonance imaging (MRI) and EEG was performed in all patients upon diagnosis.
Exclusion criteria included patients with concomitant visual impairment (visual acuity, including corrected refractory error less than 6/60), intellectual disabilities with severe developmental disorders (namely cerebral palsy, global developmental delay), history of stroke, psychiatric disorders (schizophrenia, bipolar mood disorder), and progressive neurological disorder such as (dementia, Parkinson's disease).
Assessment for visual memory was carried out using Wechsler Memory Scale-IV (WMS-IV) with scores from the subsets of visual reproduction (VR) I, II and designs I, II contributing to visual memory index (VMI) score.
VR I test was carried out by showing patients a series of five designs one at a time for 10 s each. After each design was presented, the patient had to reproduce the design immediately from memory. After 20 min, the patient would be asked to draw the designs again from memory in any order for assessment of VR II.
Assessment of design I subset was performed by showing the patient a grid with 4–8 designs on a page for 10 s, which was then removed. The patient then selected the design from a set of cards and placed the cards in a grid in the same place as previously shown. After 20 min, the patient would recreate the grid design that was shown earlier from memory. The raw score of each subset was converted to a scaled score equivalent chart based on the age group. The sum of the scaled score was then converted to the VMI score. The patient's obtained VMI score was interpreted in qualitative descriptions based on a normal curve chart in WMS-IV.
All data collected were filled in the patient data collection sheet and entered into a statistical package for social sciences (SPSS) version 25.0 for analysis. The data collected were tabulated. The correlation between continuous variables was analyzed using the Pearson correlation; whereas the VMI scores of different factors were analyzed via a 1-way ANOVA test. The statistical significance was set at P < 0.05.
Demographic and clinical characteristics
The demographic and clinical characteristics of our patients' population were summarized in [Table 1].
Prevalence of visual memory dysfunction
The prevalence of visual memory dysfunction (defined as VMI score less than 70 and qualitative description extremely low) from our study was 37.2% while 62.8% of epilepsy patients had a VMI score of 70 and above.
Factors influencing visual memory dysfunction
Analysis of individual predictors revealed that older patients, lower educational level, combined generalized and focal types of epilepsy, longer duration of epilepsy, a higher number of AEDs used, and abnormal brain imaging were predictors of poor visual memory performance.
There was a weak, negative correlation between patients' age versus VMI score [r= −0.212, n = 250, P = 0.001)] [Figure 1].
There was a significant effect of educational level on VMI score at P < 0.05 with (F (3, 246) =12.26, P < 0.001). The correlation of educational level versus VMI score showed a higher educational level had a higher mean VMI score [Table 2].
There was a significant correlation between types of epilepsy on VMI score at P < 0.05 with (F (2, 247) = 9.149, P < 0.001). Patients with combined generalized and focal epilepsy scored lowest with a mean VMI score of 58.45 ± 15.35 (extremely low), followed by patients with focal epilepsy with a mean VMI score of 75.91 ± 21.39 (borderline), and the highest mean VMI score in generalized epilepsy group with a score of 83.32 ± 20.35 (low average).
The assessment of the correlation between VMI score and duration of epilepsy showed a weak but negative correlation between these two variables with (r = −0.292, n = 250, P < 0.001). Therefore, a longer duration of epilepsy had a poorer VMI score [Figure 2].
There was a significant correlation between AEDs and VMI score at P < 0.05 with (F (5, 244) = 4.418, P = 0.001). Patients with taking higher number of AEDs performed worst [Table 3].
Abnormal imaging was defined as cortical thickness changes, hippocampal atrophy, and sclerosis from brain MRI. The VMI score was affected by abnormal brain imaging P < 0.05 for (F (1, 248) = 11.16, P = 0.001). The mean VMI score for epilepsy patients with normal imaging was 80.42 ± 21.15 (low average), while the mean VMI score for epilepsy patients with abnormal imaging was 68.57 ± 19.95 (extremely low).
There was no significant correlation between the VMI score versus abnormal EEG (presence of epileptiform wave) upon diagnosis with (F (1,248) = 1.669, P = 0.198). Age of onset of epilepsy had no association with the VMI score with (r = 0.033, n = 250, P = 0.604).
Multiple logistic regression analysis was performed to predict the VMI score from multiple factors as shown in [Table 4]. Educational level, types of epilepsy, and the number of AEDs used were factors that significantly predicted the VMI score at F (8,241) = 9.568, P < 0.001; R 2 = 0.241.
Visual memory dysfunction is associated with epilepsy with 37.2% of our epilepsy patients had visual memory dysfunction. Miller et al. reported a prevalence of visual memory impairment among older patients aged more than 55 years using Brief Visuospatial Memory Test (revised) at 39.5%; whereas a study by Nouha et al. involving patients with TLE, aged from 18 to 65 years old showed an even higher prevalence of visual memory impairment up to 78.1%.
The clinical profile of our subjects showed that increasing age correlated negatively with the VMI score significantly, which was consistent with previous literature showing that the older population with epilepsy had more marked visual memory dysfunction as compared to the control group. A local study  done in Malaysia also showed neuropsychological profiles for patients with epilepsy younger than 40 years old were significantly better as compared to the older group. This observation could be due to the concomitant neurodegenerative process in the older population.
Patients with higher educational levels also significantly had a higher VMI score in comparison to lower education levels. Younger patients and higher educational levels performed better in the visual memory test as the test required concentration, skills such as thought organization, recognition and grouping of objects that could be acquired with formal education. From a study by Pai and Tsai, at 12 months of follow-up after the diagnosis of epilepsy, patients with higher educational levels performed better in cognitive functioning. This finding suggested that training with education could compensate for cognitive decline.
Patients with combined generalized and focal types of epilepsy performed worst followed by focal epilepsy and generalized epilepsy. This was probably because our patients with combined generalized and focal epilepsy were mostly diagnosed with genetic epilepsy syndromes since childhood and likely with neurodevelopmental anomalies. Visual memory impairment was more marked in focal epilepsy as compared to generalized epilepsy as the majority of our focal epilepsy patients were due to TLE, which was related to the hippocampus where the memory processing was located.,
Longer duration of epilepsy caused more impairment in visual memory, which was observed in other reports., It could be explained that a longer duration of epilepsy represented an accumulation of more episodes of seizures leading to cerebral dysfunction. This could result in progressive structural hippocampal damage extending to the limbic structures and thinning of the cortex, thereby causing progressive memory impairment.
AEDs are still the first-line treatment for patients with epilepsy. AEDs are important to control breakthrough seizures to prevent neuronal damage. However, they also alter excitation levels of the central nervous system, which could exacerbate cognitive deficits. The use of AEDs was strongly associated with affecting memory function as shown in various studies.,, The risk of adverse cognitive side effects increased with rapid titration, higher target doses of AEDs in the blood level and a higher number of concurrent AEDs.,
Our subjects with abnormal imaging underperformed on the visual memory test. Smaller baseline white matter and hippocampal sclerosis were significant predictors of adverse effects on cognitive impairment.,, These findings were not consistent as there was no difference in memory dysfunction between negative MRI patients versus patients with hippocampal sclerosis. The involvement of other abnormal brain network functions might contribute to memory impairment.,
From our study, there was no significant correlation between abnormal EEG findings upon diagnosis and the age of onset of epilepsy on visual memory dysfunction. Factors that increased the chances of capturing epileptic discharges were recording of EEG within 48 h of the seizure onset and ongoing seizure with the frequency of at least one attack per month. The abnormal EEG findings in our study might be an underestimation as some of the EEG were performed as an outpatient setting rather than within 48 h of seizure onset. Moreover, cognitive impairment required a longer duration to develop and the risk of developing cognitive impairment is reduced once seizures were controlled. Thus, a single abnormal EEG that was performed at diagnosis was not able to predict the likelihood of visual memory impairment.
There was no significant correlation between the age of onset of epilepsy and the VMI score, which was supported by a similar study. However, there were different findings, in which patients with epilepsy presented in childhood or adolescence might affect memory. The onset of seizures at a young age might cause a reduction of neurogenesis and recurrent seizures would cause memory dysfunction.,, This was different in other populations where these studies were focused on focal epilepsy.
Assessment of epilepsy on visual memory dysfunction was influenced by multifactorial causes. Cognitive impairment might be due to the origin of the interictal epileptiform discharges, the natural history of epilepsy or negative effect from AEDs. In our study, when comparing all these contributing factors, only types of epilepsy, educational level, and usage of AEDs were found to be significant predictors.
This study was limited by the usage of two subsets of visual memory tests to determine visual memory dysfunction and there was no healthy control group. Future studies are encouraged to involve other subsets of visual memory assessment with age and gender-matched control group. Another important area that can be included in the subgroup analysis is seizure control, as poorly controlled seizures may contribute to visual memory impairment.
In conclusion, WMS-IV is a suitable assessment tool for the determination of visual memory function. Our current study enabled us to identify the prevalence of visual memory dysfunction among our patients and predictors that contributing to the impairment. This is crucial in managing our patients as it helps us to identify modifiable factors such as optimization of AEDs by slow titration of AEDs and minimizing polytherapy in patients who are at higher risk. Our report is instrumental in identifying patients at risk of developing visual memory impairment so that early referral for cognitive rehabilitation can be made.
We thank all the patients who participated in this study.
Declaration of patient consent
The authors certify that they have obtained all appropriate patient consent forms. In the form, the patient(s) has/have given his/her/their consent for his/her/their images and other clinical information to be reported in the journal. The patients understand that their names and initials will not be published and due efforts will be made to conceal their identity, but anonymity cannot be guaranteed.
Financial support and sponsorship
Conflicts of interest
There are no conflicts of interest.
[Figure 1], [Figure 2]
[Table 1], [Table 2], [Table 3], [Table 4]