Seizure 26 (2015) 65–68

Contents lists available at ScienceDirect

Seizure
journal homepage: www.elsevier.com/locate/yseiz

Review

Cognition in the early stages of adult epilepsy§

Juri-Alexander Witt, Christoph Helmstaedter *
Department of Epileptology, University of Bonn, Bonn, Germany

A R T I C L E I N F O A B S T R A C T

Article history: Purpose: The impact of duration of epilepsy on cognition has been discussed for a long time. More
Received 28 November 2014 recently, it has been recognized that cognitive deficits are often already present at the onset of epilepsy
Received in revised form 30 January 2015 or even before. From an etiological point of view it is now understood that it is not really the question
Accepted 31 January 2015
what comes first, epilepsy or cognitive comorbidity. Instead the evidence suggests that both problems
rather originate from a common underlying pathology.
Keywords: Methods: We selected studies addressing cognition in adult new-onset or newly diagnosed epilepsies
Epilepsy
before treatment initiation. Potential factors are outlined that affect cognition prior to, around or after
Adults
epilepsy onset.
Neuropsychology
Cognition Results: Most studies investigated newly diagnosed patients, but many included individuals who
Newly diagnosed already had a long history of seizures at the time of diagnosis. Fewer studies focused on new-onset
New onset epilepsies. Overall, cognitive problems in the early stages of adult onset epilepsy were found to be
common. The occurrence of seizures may initially cause greater concern and lead to an underreporting of
cognitive problems prior to or around the time of diagnosis.
Conclusion: The high prevalence of objective cognitive impairments present at epilepsy onset calls for
early neuropsychological assessments soon after the diagnosis of epilepsy, and at best before medical
treatment is initiated. Without such baseline assessments subsequent neuropsychological testing during
follow-up is difficult to interpret in regard to the effects of treatment success or the course of underlying
disease processes. Beyond that, the baseline assessment may also guide treatment choices and serve as
an early indicator of the need for support or rehabilitation. In this way neuropsychological monitoring
can improve individual medical care, and increase tolerability, adherence, and treatment retention from
the point of diagnosis.
ß 2015 British Epilepsy Association. Published by Elsevier Ltd. All rights reserved.

Cognitive deficits in epilepsy are frequent and can have first hit or onset of epilepsy must be considered as an important
negative effects on daily functioning. The etiology of cognitive moderator variable, since early epilepsy and underlying patholo-
deficits is often multi-factorial, since static and dynamic factors gies can negatively affect brain maturation and development
can synergistically affect cognition [1,2]. Static factors primarily [3,4]. Additionally, individual reserve capacities and age- and sex-
refer to the presence of developmental or acquired cerebral lesions dependent neural plasticity need to be considered [5].
causing both, epilepsy and cognitive impairment. Dynamic factors Against the background of this etiological model of cognitive
contributing to cognitive impairment are (1) active epilepsy, i.e. deficits in epilepsy, the situation apparently becomes more and
seizures and interictal epileptic discharges, (2) antiepileptic drug more complex with an increasing duration of epilepsy (see Table 1
treatment (AEDs), and (3) psychiatric comorbidities. However, for an overview of the potential factors that may affect cognition in
these factors are not necessarily independent of each other. Age at the course of time). In the later stages of chronic epilepsy, and
without repeated standardized assessment, it is hardly possible to
retrospectively attribute cognitive deficits to particular factors
§
One of the authors of this paper is a member of the current editorial team of which may be involved (cf. [6] for an attempt on a group level). This
Seizure. The supervision of the independent peer review process was undertaken emphasizes the need for neuropsychological assessments at an
and the decision about the publication of this manuscript were made by other early stage of epilepsy to be able to disentangle the complex
members of the editorial team.
interactions of the factors contributing to cognitive problems [7].
* Corresponding author at: Department of Epileptology, University of Bonn,
Ideally, cognition should be assessed shortly after the onset of
Sigmund-Freud-Str. 25, 53105 Bonn, Germany.
Tel.: +49 228 287 16108; fax: +49 228 287 90 16108. epilepsy and, at the latest, before treatment initiation. Such a
E-mail address: [email protected] (C. Helmstaedter). baseline assessment is required for a subsequent monitoring of

http://dx.doi.org/10.1016/j.seizure.2015.01.018
1059-1311/ß 2015 British Epilepsy Association. Published by Elsevier Ltd. All rights reserved.
66 J.-A. Witt, C. Helmstaedter / Seizure 26 (2015) 65–68

Table 1
Potential factors affecting cognition before and at epilepsy onset, and thereafter. Gray boxes indicate a potential impact.

Prior to At Controlled epilepsy Cured Chronic

epilepsy epilepsy (100% seizure epilepsy refractory
onset onset control) (seizure free, epilepsy
AEDs
withdrawn)
Cerebral
lesions

Covert epileptic
dysfunction

Overt epileptic
seizures

Antiepileptic
treatment

Behavioral/
psychiatric
problems

treatment success and the course of the disease. Furthermore, which needs to be considered is age at the onset of epilepsy, which
information on the cognitive status can also indicate the need for ranges on average between 27 and 71 years. The age at the onset
rehabilitative procedures, or it can direct treatment choices. By of epilepsy is strongly connected to its etiology. Therefore the
means of cognitive monitoring, neuropsychology allows for quality results obtained in early versus late onset epilepsies can hardly be
and outcome control of medical interventions striving for an compared without taking pathology and the interaction of
improved individual medical care [2]. pathology with the maturing versus aging brain into consideration.
Since the mutual relationship between pathology, epilepsy, and In the light of the increasing incidence of epilepsy with older age, a
cognition has become increasingly recognized, cognition in people recent study investigated elderly patients with new-onset epilepsy
with new-onset epilepsy has been studied more intensively. In aged between 60 and 95 years. As expected elderly patients
2011, this issue was identified as a theme of major importance for showed greater impairments than the younger sample [20].
neuropsychological research in the field of epilepsy [8]. Beforehand Some studies analyzed patients without evidence of brain
it is important to point out that a careful distinction needs to be pathology in order to disclose the sole effect of active epilepsy on
made between research relating to children and adults. In children, cognition [12,14,18]. One study considered epilepsy syndromes
neuropediatric services and parents are concerned with brain and and compared the neuropsychological results between idiopathic,
mental development as soon as a patient is diagnosed with symptomatic and cryptogenic epilepsies [19]: Among these
epilepsy. For a review focusing on cognition in pediatric patients subtypes, symptomatic epilepsies presented with the worst
with new-onset seizures please see Hermann et al. [9]. In children performance in executive function. Finally, most studies simply
it has been demonstrated how early assessment allows for the compared the cognitive performance of patients with newly
monitoring of the subsequent course of epilepsy [10,11]. This may diagnosed epilepsy and healthy controls on a group level. Only five
well serve as a model for adult epileptology. However, in adults it is of the 11 studies assessed more than 100 patients, a minority had
still common practice at the stage of diagnosis to explore the larger reference groups of healthy subjects, and only six studies
etiology of seizures and focus on rapid seizure control but not to provide prevalences/frequencies of cognitive deficits on an
take account of comorbidities such as cognitive difficulties. individual level according to normative data or healthy controls
For this review we performed a literature search (medline) for [18–20]. Taylor et al. report impaired performance in 1–18% of all
original articles investigating cognition in ‘‘untreated’’ ‘‘adult measures of the employed test battery [18]. Fifty-four percent of
patients’’ with ‘‘new-onset epilepsy’’ or ‘‘newly-diagnosed epilep- the patients versus 21% of the healthy controls were impaired in at
sy’’. Only studies with an objective cognitive assessment were least one test score. A study in 247 middle-aged patients with new-
considered. A total of 11 studies met these criteria. Most of them onset epilepsy found impairments in attention and memory in 48–
addressed at least attention and memory and reported respective 49% of the sample [19]. Less than one third was unimpaired in both
impairments in these domains [12–22] (for overview see Table 2). domains. Subjective deficits in the respective domains were
However, the studies differed considerably in regard to patient complained by 25–29% of the patients only. This indicates an
selection, sample sizes, etiologies, assessment tools, the question underreporting of cognitive deficits at this early stage of epilepsy.
of any pretreatment, and most importantly in regard to the Comparable underreporting was evident in 257 elderly patients
duration of untreated epilepsy. A point of major importance is that with new-onset epilepsy [20]: The prevalence of objective deficits
most studies investigated newly diagnosed patients and not in executive function was 58%, whereas subjective deficits
patients with new-onset epilepsies, emphasizing the need for a were reported in only up to 27% of the patients. The independent
clear terminology [23]. Indeed the mean duration of epilepsy in the observation in two studies that cognitive problems were
different patient cohorts ranged from 92 days to 7 years. Even the underreported when compared to objective assessment may be
newly diagnosed patients of the SANAD study presented with a interpreted in a way that in the patients’ view and at that early
mean epilepsy duration of 5 years [18]. The same is true for an early stage of the disease the diagnosis of seizures and their expected
study performed at our center comparing the cognitive effects of consequences are of pressing relevance. In chronic epilepsies
first treatment on lesional versus non-lesional patients [22]. This cognitive complaints are much more frequent [24,25].
means that the neuropsychological findings from such studies The finding that cognitive-behavioral deficits may precede
cannot tell us much about new-onset epilepsy. Another factor seizure onset [26], raises the question of whether there is a
 J.-A. Witt, C. Helmstaedter / Seizure 26 (2015) 65–68 67

Table 2
Neuropsychological studies in untreated patients with new-onset or newly diagnosed epilepsy.

First author Year Investigated sample N Age [M (SD)] Duration of Reference Evaluated cognitive domains Pre-treatment
epilepsy group impairment
[M (SD)]

Kälviäinen 1992 Cryptogenic epilepsies 74 32 (12) n.a.a 39 healthy Attention, memory, language, 26–39%
intelligence
Helmstaedter 1993 Symptomatic (75%), idiopathic 16 27.2 (9.0) 6.7 years 19 healthy Attention, language, memory 3/10 measures
(25%) epilepsies
Äikiä 1995 Cryptogenic epilepsies 56 34.2 (14.4) n.a. 48 healthy Verbal memory, intelligence 14–75%
Prevey 1998 Symptomatic epilepsies 201 45.7 (15.7) 7.3 years 45 healthy Attention and executive 17/18 measures
functions, memory, motor
functions, intelligence
Ogunrin 2000 Focal or generalized seizures 60 31.6 (17.4) n.a. 60 healthy Attention, memory 7/8 measures
Pulliainen 2000 Focal or generalized seizures 52 30.2 (12.6) 1 year (54%), 26 healthy Attention, memory, motor 5/20 measures
>1 year (46%)b functions
Äikiä 2001 Left temporal lobe epilepsies 39 35.3 (15.4) 3.5 (5.5) years 46 healthy Verbal memory, verbal 44–92%
(62% cryptogenic) 16 treated intelligence
patient with
epilepsy
Wesnes 2009 Focal or generalized seizures 570 29.5 (median) n.a. Test norms Attention and executive Present, but not
functions, memory, motor specified
functions, fluid intelligence
Taylor 2010 Non-lesional with focal or 155 35.0 (14.4) 5.1 years 87 healthy Attention and executive 1–18%; 6/14
generalized seizures functions, memory measures
Witt 2012 Symptomatic (48%), 247 47.0 (18.8) 92.4 (94.4) 220–359 Attention and executive 48–49%, no
cryptogenic (27%), idiopathic days healthy functions, verbal memory impairment
(25%) epilepsies at all in 28%
Witt 2014 Symptomatic (88%), 257 71.5 (7.2) <6 months 689 healthy Attention and executive 58%
cryptogenic (12%) epilepsies functions

M, mean; SD, standard deviation; n.a., not available.

a
A median of 1 generalized and 0 complex-partial seizures before inclusion.
b
In 15% first manifestation before >10 years.

bidirectional relationship between the cognitive deficits and summarizes the potential factors that affect cognition prior to
epilepsy. It is clear that people with epilepsy have a higher risk and at epilepsy onset as well as in the time after the development
of cognitive deficits. But do people with cognitive impairment also of seizures. It demonstrates, how difficult it is to isolate the
have a higher risk of developing epilepsy? Although this is rather impact of an individual factor on cognition. The most discrimina-
an epidemiological question, one may pursue a more theoretical tive information can be provided by intraindividual follow-up
approach to answer this question. First of all, epilepsy is a examinations taking account of treatment changes and variations
dysfunction of the brain associated with hyperexcitability leading in seizure control [30].
to recurrent seizures. In symptomatic epilepsies, the dysfunction is In new-onset symptomatic/structural epilepsy, the cognitive
caused by an epileptogenic lesion. Depending on its etiology and profile at epilepsy onset is primarily determined by the char-
pathology, such a lesion may exist for a long time before the acteristics of the underlying brain lesion (time of damage/onset,
occurrence of the first seizure (cf. latent period in temporal lobe extent, site, and side), eventual interictal epileptic discharges, and
epilepsy with hippocampal sclerosis). Since the lesion itself can psychiatric problems. Individual patient characteristics and
cause cognitive impairments, it would be not surprising that related differences in reserve capacities and neural plasticity need
patients with new-onset symptomatic epilepsy exhibit a cognitive to be considered as well.
deficit consistent with the localization, lateralization and extent of In idiopathic generalized epilepsies, genetically mediated
the lesion (cave: very early brain damage may additionally affect alterations of nerve cell proteins (e.g. subunits of ion channels)
development on a more global level thus causing intellectual are assumed to enhance neuronal excitability. Gross structural–
disability). Moreover, one cannot rule out that covert epileptic morphological brain abnormalities are usually not present. Here,
dysfunction affects cognition even before the first overt seizure, as cognitive problems are primarily caused by dynamic epileptic
may be observed with transitory cognitive impairment due to dysfunction, and also by secondary behavioral and psychiatric
interictal epileptic discharges [27]. Last but not least, psychiatric problems. Such problems may precede the first seizure, leading to
disturbances that negatively affect cognition may also precede or dynamic and variable cognitive deficits as compared to the more
even promote epilepsy. For instance, there are studies indicating a stable lesion-related impairments. Nevertheless, they may have,
bidirectional relationship between depression and epilepsy even if reversible, a negative impact on cognitive development in
[28,29]. In addition, it needs to be considered that the new general.
diagnosis of epilepsy may have adverse effects on the patient’s The high prevalence of objective cognitive deficits, even at the
mood and due to this also on cognitive performance. However, one onset of epilepsy, the underreporting of cognitive problems by the
study in new-onset epilepsies found no correlation between the patient, and the increasing number of factors contributing to
self-rated impact of the diagnosis on psychic well-being and cognitive dysfunction with more chronic epilepsy and its treatment
objective cognitive performance [19]. call for routine neuropsychological baseline assessments soon
In summary, the current evidence from studies in newly after the onset/diagnosis, at least before treatment is initiated.
diagnosed and new-onset epilepsies show that cognitive deficits As outlined above, such a baseline assessment is a prerequisite
are already very common at epilepsy onset. An important and for valid repeat evaluations undertaken to reflect the course of
unanswered question is whether the cognitive impairment the disease and treatment success. Moreover, the baseline assess-
resulting from acquired or developmental cerebral lesions is a ment may direct drug choice and can be taken as an early
marker for the risk of future epilepsy (or epileptogenesis). Table 1 indicator of the need of supportive or rehabilitative care.
68 J.-A. Witt, C. Helmstaedter / Seizure 26 (2015) 65–68

However, it is still very uncommon for epilepsy services to [8] McDonald CR, Taylor J, Hamberger M, Helmstaedter C, Hermann BP, Schefft B.
Future directions in the neuropsychology of epilepsy. Epilepsy Behav 2011;22:
offer baseline cognitive assessments at the time of the diagnosis 69–76.
nor subsequent cognitive monitoring as a matter of routine. [9] Hermann BP, Jones JE, Jackson DC, Seidenberg M. Starting at the beginning: the
Cognitive assessment is often not reflected in the cost structure of neuropsychological status of children with new-onset epilepsies. Epileptic
Disord 2012;14:12–21.
health care providers, and it is, at present, mostly limited to tertiary [10] Lin JJ, Dabbs K, Riley JD, Jones JE, Jackson DC, Hsu DA, et al. Neurodevelopment
epilepsy centers. Unfortunately, most patients with new-onset in new-onset juvenile myoclonic epilepsy over the first 2 years. Ann Neurol
seizures will in first instance see a general practitioner, resident 2014;76:660–8.
[11] Rathouz PJ, Zhao Q, Jones JE, Jackson DC, Hsu DA, Stafstrom CE, et al. Cognitive
neurologist or an acute care hospital without any or only development in children with new onset epilepsy. Dev Med Child Neurol
limited and non-specialized neuropsychological service. One possi- 2014;56:635–41.
ble solution would be to establish clinics specializing in new- [12] Aikia M, Kalviainen R, Riekkinen PJ. Verbal learning and memory in newly
diagnosed partial epilepsy. Epilepsy Res 1995;22:157–64.
onset epilepsies which offer sufficient neuropsychological expertise
[13] Äikiä M, Salmenperä T, Partanen K, Kälviäinen R. Verbal memory in newly
to provide baseline and early follow-up assessments (for example diagnosed patients and patients with chronic left temporal lobe epilepsy.
as an extension of tertiary epilepsy centers). An alternative Epilepsy Behav 2001;2:20–7.
would be to improve the conditions of ambulatory neuropsycho- [14] Kälviäinen R, Aikiä M, Helkala EL, Mervaala E, Riekkinen PJ. Memory and
attention in newly diagnosed epileptic seizure disorder. Seizure 1992;1:
logical services. The actual situation, at least in Germany, is quite 255–62.
disappointing, both in terms of financing and with regard to [15] Ogunrin O, Adamolekun B, Ogunniyi AO, Aldenkamp AP. Cognitive function in
professional policy [31]. Nigerians with newly diagnosed epilepsy. Can J Neurol Sci 2000;27:148–51.
[16] Prevey ML, Delaney RC, Cramer JA, Mattson RH. Complex partial and second-
Neuropsychology can provide valid and reliable tools for quality arily generalized seizure patients: cognitive functioning prior to treatment
and outcome control of medical interventions. This has been with antiepileptic medication. VA Epilepsy Cooperative Study 264 Group.
demonstrated particularly in the context of epilepsy surgery but Epilepsy Res 1998;30:1–9.
[17] Pulliainen V, Kuikka P, Jokelainen M. Motor and cognitive functions in newly
also in the context of medical treatment and pharmaceutical drug diagnosed adult seizure patients before antiepileptic medication. Acta Neurol
trials. Disseminating this potential across institutions and coun- Scand 2000;101:73–8.
tries remains a major challenge. European initiatives such as E- [18] Taylor J, Kolamunnage-Dona R, Marson AG, Smith PE, Aldenkamp AP, Baker
GA. Patients with epilepsy: cognitively compromised before the start of
Brain (www.ebrainjnc.com), or European networks like E-PILEPSY antiepileptic drug treatment? Epilepsia 2010;51:48–56.
(www.e-pilepsy.eu/) which aim at a harmonization of diagnostic [19] Witt JA, Helmstaedter C. Should cognition be screened in new-onset epilep-
procedures and a common web-based platform for European sies? A study in 247 untreated patients. J Neurol 2012;259:1727–31.
[20] Witt J-A, Werhahn KJ, Krämer G, Ruckes C, Trinka E, Helmstaedter C. Cognitive-
epilepsy centers may represent a first important step towards a
behavioral screening in elderly patients with new-onset epilepsy before
more routine use of neuropsychological assessment. treatment. Acta Neurol Scand 2014;130:172–7.
[21] Wesnes KA, Edgar C, Dean AD, Wroe SJ. The cognitive and psychomotor effects
of remacemide and carbamazepine in newly diagnosed epilepsy. Epilepsy
Conflict of interest statement
Behav 2009;14:522–8.
[22] Helmstaedter C, Wagner G, Elger CE. Differential effects of first antiepileptic
None of the authors has any conflict of interest in regard to the drug application on cognition in lesional and non-lesional patients with
topic and the contents of this article. epilepsy. Seizure 1993;2:125–30.
[23] Pohlmann-Eden B, Legg K, Crocker CE. Definition of new-onset epilepsy versus
newly diagnosed epilepsy: role of time domain. Epilepsia 2012;53:1277.
References [24] Hoppe C, Elger CE, Helmstaedter C. Long-term memory impairment in patients
with focal epilepsy. Epilepsia 2007;4:26–9.
[1] Elger CE, Helmstaedter C, Kurthen M. Chronic epilepsy and cognition. Lancet [25] Helmstaedter C. Neuropsychological complaints, deficits and difficulties in
Neurol 2004;3:663–72. every day life. In: Pfäfflin RTF, Thorbecke R, Specht U, Wolf P, editors.
[2] Helmstaedter C, Witt JA. Clinical neuropsychology in epilepsy – theoretical and Comprehensive care for people with epilepsy. London: John Libbey & Compa-
practical issues. In: Theodore W, Stefan H, editors. Handbook of clinical ny Ltd.; 2001. p. 293–306.
neurology (107). Epilepsy. Part I: Basic principles and diagnosis. Elsevier; 2012. [26] Austin JK, Harezlak J, Dunn DW, Huster GA, Rose DF, Ambrosius WT. Behavior
[3] Helmstaedter C, Elger CE. Chronic temporal lobe epilepsy: a neurodevelop- problems in children before first recognized seizures. Pediatrics 2001;107:
mental or progressively dementing disease? Brain 2009;132:2823–30. 115–22.
[4] Kaaden S, Quesada CM, Urbach H, Koenig R, Weber B, Schramm J, et al. Neu- [27] Binnie CD. Cognitive impairment during epileptiform discharges: is it ever
rodevelopmental disruption in early-onset temporal lobe epilepsy: evidence justifiable to treat the EEG? Lancet Neurol 2003;2:725–30.
from a voxel-based morphometry study. Epilepsy Behav 2011;20:694–9. [28] Kanner AM. Depression and epilepsy. A bidirectional relation? Epilepsia
[5] Helmstaedter CA. Prediction of memory reserve capacity. Adv Neurol 1999;81: 2011;52(Suppl. 1):21–7.
271–9. [29] Kanner AM. Can neurobiological pathogenic mechanisms of depression facili-
[6] Witt J-A, Elger CE, Helmstaedter C. Impaired verbal fluency under topiramate – tate the development of seizure disorders? Lancet Neurol 2012;11:1093–102.
evidence for synergistic negative effects of epilepsy, topiramate, and poly- [30] Witt JA, Helmstaedter C. Monitoring the cognitive effects of antiepileptic
therapy. Eur J Neurol 2013;20:130–7. pharmacotherapy - approaching the individual patient. Epilepsy Behav
[7] Helmstaedter C, Aldenkamp AP, Baker GA, Mazarati A, Ryvlin P, Sankar R. 2013;26:450–6.
Disentangling the relationship between epilepsy and its behavioral comorbid- [31] Herrmann M, Willmes-von Hinckeldey K, Sturm W, Wallesch C-W, Karnath H-
ities – the need for prospective studies in new-onset epilepsies. Epilepsy Behav 132?O, Jäncke L, et al. Klinische neuropsychologie ist keine psychologische
2014;31:43–7. psychotherapie. Z Neuropsychol 2014;25:65–8.