Is long-term electroencephalogram more appropriate than standard electroencephalogram in the elderly?

Clinical Neurophysiology xxx (2016) xxx–xxx

Contents lists available at ScienceDirect

Clinical Neurophysiology journal homepage: www.elsevier.com/locate/clinph

Is long-term electroencephalogram more appropriate than standard electroencephalogram in the elderly? M. Chochoi a, L. Tyvaert a, P. Derambure a,b, W. Szurhaj a,b,⇑ a b

Epilepsy Unit, Lille University Hospital, Lille, France INSERM U1171, Univ. Lille, F-59000 Lille, France

a r t i c l e

i n f o

Article history: Received 21 February 2016 Revised 1 October 2016 Accepted 8 October 2016 Available online xxxx Keywords: All epilepsy/seizures Non convulsive seizures Elderly Electroencephalogram

h i g h l i g h t s  Standard EEG has a low diagnostic yield in elderly patients with suspected non-convulsive seizures.  Long-term EEG was significantly more sensitive than standard EEG in detecting epileptiform

discharges.  Epileptiform discharges were often recorded during sleep only.

a b s t r a c t Objective: To compare the diagnostic accuracy of standard (st) and long-term video (lt) EEG in elderly patients with suspected non-convulsive seizures. Methods: Over a 12-month period, we prospectively included all elderly (over-65) hospitalized patients having undergone lt-EEG for suspected non-convulsive seizures (n = 43). st-EEG was defined as the first 20 min of each lt-EEG. We recorded the patients’ clinical and imaging characteristics and final diagnosis and assessed the respective diagnostic values of st-EEG and lt-EEG. Results: Epileptiform discharges were detected on standard EEG in only 7% of patients and in 28% of patients on Lt-EEG (p = 0.004). Non-convulsive seizures were recorded in 1 case vs. 4, respectively. Nine of 40 negative standard EEG showed later epileptiform activities. The median time to occurrence of the first epileptiform activities was 46.5 min (interquartile range: 36.5–239.75 min). Epileptiform activity occurred during sleep only in 33% patients with a negative st-EEG. Dementia was associated with a positive lt-EEG (p:0.047). Conclusion: Lt-EEG was clearly superior to standard EEG for detecting epileptiform activity in elderly when suspecting non convulsive seizures. Significance: St-EEG has a low diagnostic yield in elderly patients with suspected non-convulsive seizures and so lt-EEG is preferable in this situation. Ó 2016 International Federation of Clinical Neurophysiology. Published by Elsevier Ireland Ltd. All rights reserved.

1. Introduction Epilepsy is frequent in the elderly; the estimated prevalence is 1–2% in people over the age of 60 (Brodie and Kwan, 2005; Abbreviations: EEG, electroencephalogram; st-EEG, standard electroencephalogram; lt-EEG, long-term electroencephalogram; IEDs, interictal epileptiform discharges; MRI, magnetic resonance imaging. ⇑ Corresponding author at: Department of Clinical Neurophysiology, Hopital Salengro, Rue Emile Laine, 59037 Lille Cedex, France. Fax: +33 320446355. E-mail addresses: [email protected] (M. Chochoi), [email protected] (L. Tyvaert), [email protected] (P. Derambure), [email protected] (W. Szurhaj).

Hauser et al., 1993) and 7.7% in institutionalized patients over 65 (Garrard et al., 2003). The increasing incidence of epilepsy in the elderly has been linked to the increase in ‘‘structural” causes – especially stroke (So et al., 1996) and dementia. However, around 25% of cases are ‘‘idiopathic” (Ramsay et al., 2004), which suggests that the brain is predisposed to generate epileptic seizures. In elderly people, the diagnosis of seizures is often complicated by the fact that the manifestations may be subtle or perplexing (Silveira et al., 2011; Brodie et al., 2009; Oono et al., 2014). Moreover, the patient interview may be complicated by cognitive disorders and the absence of witnesses. Hence, other diagnoses are sometimes considered before epilepsy (Chernyshev et al., 2010;

http://dx.doi.org/10.1016/j.clinph.2016.10.006 1388-2457/Ó 2016 International Federation of Clinical Neurophysiology. Published by Elsevier Ireland Ltd. All rights reserved.

Please cite this article in press as: Chochoi M et al. Is long-term electroencephalogram more appropriate than standard electroencephalogram in the elderly?. Clin Neurophysiol (2016), http://dx.doi.org/10.1016/j.clinph.2016.10.006

2

M. Chochoi et al. / Clinical Neurophysiology xxx (2016) xxx–xxx

Shavit et al., 2012). However, it is essential to diagnose and treat epilepsy promptly because of the possible complications of seizures and because the over-diagnosis of epilepsy unnecessarily exposes often fragile elderly patients to adverse drug reactions (Besocke et al., 2013). Epilepsy may contribute to the aggravation of dementia, and may predispose stroke patients to dementia (Cordonnier et al., 2007). EEG seems not to be a very useful diagnostic tool in the elderly. Even if some studies show the same sensibility and sensitivity of EEG in elderly than in younger patients (Watson et al., 2012), most of studies are in favour of lower specificity and sensibility in elderly. Indeed, only 26% of over-60 patients with epilepsy reportedly display interictal epileptiform discharges (IEDs) (Drury and Beydoun, 1998). Hence, EEG’s low sensitivity means that the absence of epileptiform abnormalities (i.e. a negative recording) does not rule out a final diagnosis of epilepsy. In view of these findings, Epilepsy Scotland recommends that EEG should not be used to prove or disprove a diagnosis of epilepsy in the elderly (Brodie et al., 2009). Nevertheless, EEG is a cheap, easy-to-use tool. The technique therefore needs to be adapted to address this challenging, important issue in elderly people with suspected nonconvulsive seizures. Some researchers have shown that long term EEG (lt-EEG) could be very helpful to diagnose non-convulsive seizures in the elderly, especially in case of delirium (Naeije et al., 2012, 2014). It has been suggested that lt-EEG is more sensitive than standard EEG (st-EEG) for detecting interictal epileptiform abnormalities. Indeed, we hypothesized that lt-EEG might be a valuable alternative to stEEG in the elderly (especially to make unlikely a diagnosis of epilepsy). However, literature data on this interesting question are scarce. We therefore performed a prospective, comparative study of the diagnostic value of lt-EEG and st-EEG in elderly people with suspected non-convulsive seizures.

2. Methods 2.1. Inclusion and non-inclusion criteria Over a 12-month period, we prospectively included all patients meeting the following criteria: aged 65 or over; hospitalization in the Neurology Department at Lille University Hospital (Lille, France); lt-EEG recording requested by the attending physician; and suspected non-convulsive seizures. The main exclusion criteria were as follows: cancellation of the initial request for lt-EEG prior to recording; excessive agitation; a clear diagnosis after st-EEG (only 1 patient with an EEG displaying a metabolic encephalopathy), making lt-EEG unnecessary.

For each patient, we recorded the gender, age, medical history, current medications, the reason for referral, the time interval between the appearance of suspected non-convulsive seizures and the EEG, the results of the initial clinical examination, the brain imaging results, the presence of any acute disorders (recent stroke, metabolic disorders, infections, etc.), and the final diagnosis. Moreover, when patients were followed up in our hospital, we documented the diagnosis of epilepsy at subsequent visits (up until December 31st 2015). We classified the reasons for referral into as (i) an altered state or loss of consciousness, (ii) confusion or a behavioural disorder, and (iii) focal neurological signs. All epilepsy diagnoses were reviewed by a panel of three epileptologists (MC, PD and WS). The final diagnosis was categorized as (i) confirmed epilepsy, with well-documented seizures, (ii) possible epilepsy (i.e. epilepsy remained a potential diagnosis) and (iii) a clearly non-epileptic condition (another aetiology for referral was found, the events occurred during the lt-EEG in the absence of concomitant electric seizures, and/or epileptiform manifestations did not recur during follow-up). The present non-interventional study was registered with the French National Data Protection Commission (Commission Nationale de l’Informatique et des Libertés; reference: DEC2015-117). 2.3. Data analysis Each st- or lt-EEG recording was assessed by one of three expert neurophysiologists (PD, LT and WS). The neurophysiologist assessed the st-EEG recording and then the lt-EEG recording. The st-EEG and lt-EEG results were classified into four groups: – A: no focal abnormalities. – B: slow focal activities. – C: epileptiform discharges (spikes, spikes-and-waves, polyspikes, polyspikes-and-waves, or sharp waves). – D: seizures. If a given patient showed several different types of EEG activities (e.g. both B and D), only the most severe classification (D, in that case) was taken into account. Class A and B results were considered to be ‘‘negative”. Class C and D results were considered to be ‘‘positive”. The time to occurrence of the first epileptiform activities (IEDs or seizure) was determined. Lastly, the EEG’s overall impact on patient management was evaluated by the medical staff. St- or lt-EEG was considered to be useful if it influenced the patient’s management (e.g. by leading to a change in treatment or by confirming or ruling out a diagnosis of epilepsy). 2.4. Statistical analysis

2.2. Data collection Ò

All EEGs were performed with SystemPLUS equipment and software (Micromed S.p.A., Mogliano Veneto, Italy), using 19 disc electrodes attached to the scalp (according to the international 10–20 system) with conductive paste, a sampling frequency of 256 Hz, a 0.5–70 Hz band pass filter. All EEG sessions were performed at the bedside in the clinical department. In the presence of a technician, each recording started with a 20-min period that was subsequently defined as the st-EEG. This included periods of eye-opening and eye-closing, intermittent photic stimulation, hyperventilation, (unless these procedures were ruled out for medical reasons or by the patient’s inability or unwillingness to cooperate). The attending medical staff were asked to note all significant events (seizures, paroxysmal neurological events, patient care procedures, etc.)

Performance parameters for standard and lt-EEG comparison were compared in McNemar’s test. A non-parametric Mann–Whitney U test was used for intergroup comparisons of continuous variables. A chi2 test or Fisher’s test was used for intergroup comparisons of categorical variables. The threshold for statistical significance was set to p < 0.05. 3. Results 3.1. The study population Forty-three patients were included (16 men and 27 women; gender ratio: 0.6; mean ± standard deviation (range) age: 82.1 ± 6.59 (68–95)). Twenty-two of the 43 patients (51.2%) had a history of neurological disease; there were 11 cases of dementia

Please cite this article in press as: Chochoi M et al. Is long-term electroencephalogram more appropriate than standard electroencephalogram in the elderly?. Clin Neurophysiol (2016), http://dx.doi.org/10.1016/j.clinph.2016.10.006

3

M. Chochoi et al. / Clinical Neurophysiology xxx (2016) xxx–xxx

(25.6%), 10 cases of stroke (23.3%) and 1 brain tumor (2.3%). Five patients took anti-epileptic drugs (11.6%). None of the 43 patients had normal MRI results (although 2 patients only underwent a CTscan). The main imaging abnormalities were cortical and subcortical atrophy (n = 28; 65.1%), vascular leukoencephalopathy (n = 21; 48.8%), and recent stroke (n = 10; 23.3%) Thirteen patients did not have obvious aetiologies for their confused state. Other patients presented with infections (n = 8; 18.6%), metabolic disorders (n = 7; 16.3%), a recent stroke (n = 10; 23.3%), or other potential causes (a brain tumor, cardiac rhythm disorders, etc.). The reasons for suspecting non-convulsive seizures were as follows: loss of consciousness (n = 10; 23.3%), a confused state (n = 10; 23.3%), a focal neurological sign (n = 20; 46.6%) and behavioural disorders (n = 3; 7%). In 16 patients (37.2%), these signs were still present at the start of the EEG recording. Twenty-two patients were reviewed after discharge from hospital, with a mean follow-up period of 23.3 months ±12.2.

3.2. EEG results The EEG recording was started within 24 h of symptom onset in 16 cases (37.2%), between 24 and 72 h afterwards in 15 cases (34.9%) and more than 72 h afterwards in 12 cases (27.9%). Twenty-nine continuous EEG recordings (67.4%) lasted for 24 h. Fourteen EEG were shorter because patients with delirium had removed their electrodes or because of a loss of quality of signal (loss of the ground electrode or reference electrode the most often). Six EEG (14%) lasted between 12 and 24 h (including sleep), and 8 (18.6%) lasted between 3 and 12 h (with no sleep) because the patient had removed the electrodes. The EEG results are summarized in Fig. 1 and Table 1. The st-EEG recordings evidenced slow abnormalities in 25 of the patients (58%) and epileptiform activities in only 3 (7%, included one recording of a non-convulsive seizure (2%)). The ltEEG recordings revealed epileptiform activities in 12 patients (28%; p = 0.004 in a McNemar test vs. st-EEG). Nine of the 40 patients (22.5%) with a ‘‘negative” st-EEG showed epileptiform activities in the lt-EEG. In the 3 patients with a ‘‘positive” st-EEG, the lt-EEG provided additional diagnostic data by evidencing a seizure in 2 patients (including 8 seizures in one of them). The median time to occurrence of the first epileptiform activities in the whole population was 46.5 min (interquartile range: 36.5–239.75 min). The abnormalities occurred within the first hour in 6 cases (66.7%), after between 1 and 5 h in 1 case (11.1%), and after 5 h in 2 cases (22.2%). In 3 of these 9 patients (33.3%), abnormalities were only observed during sleep.

Table 1 Comparison of the results of st-EEG and Lt-EEG. St-EEG

Lt-EEG

40 ‘‘normal”/‘‘focal slowing”

31 ‘‘normal”/‘‘focal slowing” 7 ‘epileptiform discharges” 2 ‘‘seizures” 1 ‘‘epileptiform discharges” 1 ‘‘seizure” 1 ‘‘seizure” (8 seizures...)

2 ‘‘epileptiform discharges” 1 ‘‘seizure”

In 14 cases, the medical staff considered that lt-EEG was more useful than st-EEG. In 8 of these cases (18.6%), the presence of epileptiform activities (along with clinical criteria), argued in favour of a diagnosis of an epilepsy. In 6 cases (14%), even a negative lt-EEG was valuable because it made epilepsy an unlikely cause of the suspect event. 3.3. Comparison of the EEG results and the final diagnosis At last follow-up, epilepsy was the final diagnosis in 10 cases. Lt-EEG had evidenced epileptiform activities in 5 of these patients (50%). Hence, there were 5 false negatives (50%) (Table 2). A diagnosis of epilepsy was not considered in 18 cases, either because ltEEG did not reveal concomitant ictal discharges during the neurological events (in 6 cases) or because another (reversible) aetiology was found (in 12 cases). In the group of ‘possible epilepsy’ (15 patients), lt-EEG displayed epileptiform discharges in only 5 patients (33%). To note, that lt-EEG displayed epileptiform discharges in 2 of 18 patients (11.1%) in whom epilepsy was finally considered as unlikely. 3.4. Comparison of groups with and without epileptiform activities on lt-EEG Subgroups of patients with and without epileptiform activities on lt-EEG did not differ significantly in terms of the time interval between symptom onset and the start of the EEG recording or the duration of the lt-EEG (Table 3). The only significance intergroup difference was a higher frequency of dementia in the subgroup of patients with epileptiform activities. When considering patients with confusion, lt-EEG revealed non-convulsive seizures in only 1 case. 4. Discussion We prospectively evaluated lt-EEG and st-EEG as diagnostic tools in elderly people with suspected non-convulsive seizures. Standard EEG was of limited value in the present study, and lt-term EEG was clearly more sensitive for detecting IEDs and

Fig. 1. Results of St-EEG and Lt-EEG recordings. A: normal; B: slow focal activities; C: interictal epileptiform discharges. D: seizures.

Please cite this article in press as: Chochoi M et al. Is long-term electroencephalogram more appropriate than standard electroencephalogram in the elderly?. Clin Neurophysiol (2016), http://dx.doi.org/10.1016/j.clinph.2016.10.006

4

M. Chochoi et al. / Clinical Neurophysiology xxx (2016) xxx–xxx

Table 2 Results of st-EEG and lt-EEG recordings, as a function of the final diagnosis. Results

A+B

C

D

Type of EEG

St-EEG

Lt-EEG

St-EEG

Lt-EEG

St-EEG

Lt-EEG

Epilepsy (n = 10) Possible epilepsy (n = 15) Epilepsy non retained (n = 18)

8 14 18

5 10 16

1 1 0

1 5 2

1 0 0

4 0 0

A + B: normal or slow focal abnormalities. C: epileptiform discharges. D: seizures. To note that standard-EEG displayed epileptiform activities in only 2 of 10 patients with epilepsy as final diagnosis.

Table 3 Comparison of features of the subgroups according to the results of the lt-EEG. Features

‘‘Negative” lt-EEG (A, B) (31 patients)

‘‘Positive” Lt-EEG (C, D) (12 patients)

p

Age (years) Gender ratio Medical history

– Stroke – Known dementia – Anti-epileptic drug

81.7 ± 6.6 0.63 5 (16.1%) 5 (16.1%) 4 (12.9%)

83.2 ± 6.8 0.5 5 (41.7%) 6 (50%) 1 (8.3%)

0.385 1.000 0.11 0.047⁄ 1.000

Clinical presentation

– Loss of consciousness – Confusion/behavioural disorders – Focal neurological deficit

7 (22.6%) 9(29%) 15 (48.4%)

3 (25%) 4 (33.4%) 5 (41.7%)

0.698 0.745 0.184

MRI abnormalities

– – – –

Cortical and subcortical atrophy Vascular leukoencephalopathy Vascular scars Recent stroke

13 (41.94%) 20 (64.52%) 8 (25.81%) 8 (25.81%)

8 (66.67%) 8 (66.67%) 5 (41.67%) 1(8.33%)

0.146 1.000 0.460 0.405

Time to lt-EEG

– Persistence of neurological signs – Before 72 hours – After 72 hours

12 (38.71%) 11 (35.48%) 8 (25.81%)

4 (33.33%) 4 (33.33%) 4 (33.33%)

1.000 1.000 0.711

20.5 ± 7.1

18.5 ± 9.2

0.464

Duration of lt-EEG (in hours) ⁄

Statistically significant.

seizures in this context. Lt-EEG was judged by the attending physician to be more useful than standard EEG in around one third of the cases (14 out of 43, including 6 ‘‘negative” cases in which epileptiform activity was not detected). Although epilepsy is common (and treatments are often effective), early, accurate diagnosis in the elderly may be difficult (Berg, 2012). Even though a proportion of cases are probably missed, over-diagnosis is also a problem. Drury et al.’s study of 18 over-60 patients admitted to an epilepsy monitoring unit for paroxysmal neurological manifestations found that 8 patients without epilepsy had been treated with anti-epileptic drugs (Drury et al., 1999). In a video EEG study by McBride et al. 27 of the 73 patients were diagnosed as having non-epileptic conditions (McBride et al., 2002). Both of these studies examined patients admitted to hospital from home in a non-emergency setting. In contrast, the present study assessed acute-phase, hospitalized patients at the bedside (but not only critically ill patients). Our results confirmed the value of lt-EEG recordings and additional information provided by concomitant video recording (Abubakr and Wambacq, 2005). Our results demonstrate that st-EEG is of limited value for detecting or monitoring epilepsy in the elderly, since only 3 of the 43 recordings were ‘‘positive”. Furthermore, our results confirm a number of recommendations in the literature, and emphasize the need to adapt the way EEG is used in the elderly (Brodie et al., 2009; Smith, 2005). Our present results confirm the value of lt-EEG in the elderly. Even when st-EEG reveals IEDs, lt-EEG could still be of interest – especially for detecting non-convulsive seizures. However, the incidence of IEDs was very low in our group. In two Belgian studies, IEDs or seizures were found in 44% and 66% of lt-EEG recording (Naeije et al., 2012, 2014). However, both studies focused on patients with acute confusion, which contrasts with the clinical presentations in our study. In the subgroup of patients with acute

confusion (n = 13), 31% of the lt-EEG recordings (4 out of 13) displayed epileptiform activities, and only 1 (7%) revealed nonconvulsive seizures. However, the sample sizes were too small to be statistically significant. Nevertheless, it is likely that the populations in the above-mentioned studies differed slightly from the present study population. In the study by Drury et al. (1998), IEDs were detected in 18 of the 70 patients (26%) with epilepsy onset after the age of 60. However, the population in Drury et al.’s study differed from our study population. Interestingly, our results are consistent with those of a Scottish study of a similar population (Watson et al., 2012). In our study, patients with IEDs on the lt-EEG had significantly greater levels of cognitive impairment; indeed, this was the only significant clinical difference between patients with a ‘‘positive” lt-EEG and those with a ‘‘negative” lt-EEG. This is likely to reflect the greater susceptibility to epileptic seizures in dementia (Imfeld et al., 2013). However, this may also reflect that epilepsy in the elderly may cause dementia that is largely reversible with seizure control (Høgh et al., 2002; Ito et al., 2009). Hence, lt-EEG may be of particular value in demented patients. Calculating the exact diagnostic sensitivity and specificity of ltEEG in the present study was problematic because a definite diagnosis of epilepsy is very difficult to confirm in the elderly, and because the sample size of our work is relatively small. If we take account of well documented cases of epilepsy only (n = 10, a small sample), the sensitivity was 50%. Nevertheless, it is clear that a normal lt-EEG in patients cannot rule out a diagnosis of epilepsy. Indeed, five patients with a normal lt-EEG were ultimately diagnosed with epilepsy (with well-documented seizures). Thus, in the absence of recorded events, a normal lt-EEG must be interpreted with caution. However, the technique’s specificity appears to be reasonable: in the group of patients without epilepsy, only 2 of 18 patients (11.1%) had IEDs on their lt-EEG. This is consistent with previous work on a similar population, within which 14% of patients without epilepsy displayed IEDs (Watson et al., 2012).

Please cite this article in press as: Chochoi M et al. Is long-term electroencephalogram more appropriate than standard electroencephalogram in the elderly?. Clin Neurophysiol (2016), http://dx.doi.org/10.1016/j.clinph.2016.10.006

M. Chochoi et al. / Clinical Neurophysiology xxx (2016) xxx–xxx

Moreover, it seems that a negative lt-EEG can be as useful as a positive lt-EEG. Indeed, 6 of the 31 negative lt-EEGs were judged to be useful by the medical team. In confused elderly patients, normal st-EEG (or EEG showing slowing of the background activity) cannot rule out a diagnosis of epilepsy because prolonged post-ictal signs may be present. However, if the lt-EEG remains normal during a symptomatic period, a diagnosis of epilepsy can be ruled out with a good degree of confidence. Moreover, lt-EEG is more valuable than st-EEG for ruling out the occurrence of non-convulsive seizures. It is sometimes as important to rule out a diagnosis of epilepsy as to confirm it – especially in fragile patients. Indeed, long-term video EEG recording increases the probability of detecting clinical manifestations and thus determining whether the latter’s are epileptic or non-epileptic. Long-term EEG may avoid the inappropriate prescription of anti-epileptic drugs and the inappropriate repetition of st-EEG recordings. Most EEG abnormalities occurred within the first hour of recording. Even if all the lt-EEG did not last 24 h, (it may be difficult to obtain a prolonged lt-EEG in patients with delirium or dementia), we can assume that a 24-h recording is probably not necessary. It has already been suggested that a few hours of EEG monitoring would be of value (Naeije et al., 2012). However, it is noteworthy that abnormalities occurred only during sleep in 5 cases; hence, we consider that recordings of a few hours should include a period of sleep. 5. Conclusion The present prospective study confirms the low diagnostic yield of st-EEG in elderly patients with suspected non-convulsive seizures and highlights the diagnostic value of lt-EEG (even when negative). Twenty minutes-EEG is clearly too short to diagnose epilepsy in this population. Although long-term video EEG is labour-intensive, use of this technique for several hours (including a period of sleep) appears to be preferable in elderly subjects – especially those with cognitive impairment. Conflict of interest statement No conflict. Acknowledgments The authors thank David Fraser for comments on the manuscript’s English and the FFRE (Federation Française pour la Recherche sur l’Epilepsie) for their support to this research.

5

References Abubakr A, Wambacq I. Seizures in the elderly: video/EEG monitoring analysis. Epilepsy Behav 2005;7:447–50. Berg AT. Epilepsy is common in the elderly, but where does it go? Neurology 2012;78:444–5. Besocke AG, Rosso B, Cristiano E, Valiensi SM, del García C, Gonorazky SE, Romano LM. Outcome of newly-diagnosed epilepsy in older patients. Epilepsy Behav 2013;27:29–35. Brodie MJ, Kwan P. Epilepsy in elderly people. BMJ 2005;331:1317–22. Brodie MJ, Elder AT, Kwan P. Epilepsy in later life. Lancet Neurol 2009;8:1019–30. Chernyshev OY, Martin-Schild S, Albright KC, Barreto A, Misra V, Acosta I, Grotta JC, Savitz SI. Safety of tPA in stroke mimics and neuroimaging-negative cerebral ischemia. Neurology 2010;74:1340–5. Cordonnier C, Hénon H, Derambure P, Pasquier F, Leys D. Early epileptic seizures after stroke are associated with increased risk of new-onset dementia. J Neurol Neurosurg Psychiatry 2007;78:514–6. Drury I, Beydoun A. Interictal epileptiform activity in elderly patients with epilepsy. Electroencephalogr Clin Neurophysiol 1998;106:369–73. Drury I, Selwa LM, Schuh LA, Kapur J, Varma N, Beydoun A, Henry TR. Value of inpatient diagnostic CCTV-EEG monitoring in the elderly. Epilepsia 1999;40:1100–2. Garrard J, Harms S, Hardie N, Eberly LE, Nitz N, Bland P, Gross CR, Leppik IE. Antiepileptic drug use in nursing home admissions. Ann Neurol 2003;54:75–85. Hauser WA, Annegers JF, Kurland LT. Incidence of epilepsy and unprovoked seizures in Rochester, Minnesota: 1935–1984. Epilepsia 1993;34:453–68. Høgh P, Smith SJ, Scahill RI, Chan D, Harvey RJ, Fox NC, Rossor MN. Epilepsy presenting as AD: neuroimaging, electroclinical features, and response to treatment. Neurology 2002;58:298–301. Imfeld P, Bodmer M, Schuerch M, Jick SS, Meier CR. Seizures in patients with Alzheimer’s disease or vascular dementia: a population-based nested casecontrol analysis. Epilepsia 2013;54:700–7. Ito M, Echizenya N, Nemoto D, Kase M. A case series of epilepsy-derived memory impairment resembling Alzheimer disease. Alzheimer Dis Assoc Disord 2009;23:406–9. McBride AE, Shih TT, Hirsch LJ. Video-EEG monitoring in the elderly: a review of 94 patients. Epilepsia 2002;43:165–9. Naeije G, Gaspard N, Depondt C, Pepersack T, Legros B. Acute confusional state of unknown cause in the elderly: a study with continuous EEG monitoring. Epilepsy Behav 2012;23:330–4. Naeije G, Depondt C, Meeus C, Korpak K, Pepersack T, Legros B. EEG patterns compatible with nonconvulsive status epilepticus are common in elderly patients with delirium: a prospective study with continuous EEG monitoring. Epilepsy Behav 2014;36:18–21. Oono M, Uno H, Umesaki A, Nagatsuka K, Kinoshita M, Naritomi H. Severe and prolonged ictal paresis in an elderly patient. Epilepsy Behav Case Rep 2014;2:105–7. Ramsay RE, Rowan AJ, Pryor FM. Special considerations in treating the elderly patient with epilepsy. Neurology 2004;62:S24–9. Shavit L, Grenader T, Galperin I. Nonconvulsive status epilepticus in elderly a possible diagnostic pitfall. Eur J Intern Med 2012;23:701–4. Silveira DC, Jehi L, Chapin J, Krishnaiengar S, Novak E, Foldvary-Schaefer N, Najm I. Seizure semiology and aging. Epilepsy Behav 2011;20:375–7. Smith SJM. EEG in the diagnosis, classification, and management of patients with epilepsy. J Neurol Neurosurg Psychiatry 2005;76(Suppl. 2):ii2–7. So EL, Annegers JF, Hauser WA, O’Brien PC, Whisnant JP. Population-based study of seizure disorders after cerebral infarction. Neurology 1996;46:350–5. Watson P, Conroy A, Moran G, Duncan S. Retrospective study of sensitivity and specificity of EEG in the elderly compared with younger age groups. Epilepsy Behav 2012;25:408–11.

Please cite this article in press as: Chochoi M et al. Is long-term electroencephalogram more appropriate than standard electroencephalogram in the elderly?. Clin Neurophysiol (2016), http://dx.doi.org/10.1016/j.clinph.2016.10.006