Epilepsy & Behavior 148 (2023) 109481

Contents lists available at ScienceDirect

Epilepsy & Behavior

journal homepage: www.elsevier.com/locate/yebeh

The impact of epilepsy and antiseizure medications on sleep: Findings

from a large European survey in adults with epilepsy and matched
controls
Charlotte Lawthom a, Adrien Didelot b, Antonietta Coppola c, Ángel Aledo-Serrano d, Barbara Fazekas e,
Ricardo Sainz-Fuertes e, Adam Strzelczyk f,⇑
a
Department of Neurology, Aneurin Bevan University Health Board, Newport, UK
b
Department of Neurology, Centre Hospitalier Saint Joseph Saint Luc, Lyon, France
c
Epilepsy Centre, Department of Neuroscience, Odontostomatological and Reproductive Sciences, Federico II University of Naples, Naples, Italy
d
Epilepsy Unit, Vithas Neuroscience Institute, La Milagrosa University Hospital, Madrid, Spain
e
Eisai Europe Ltd, Hatfield, Hertfordshire, UK
f
Epilepsy Center Frankfurt Rhine-Main, Department of Neurology, Goethe University and University Hospital Frankfurt, Frankfurt am Main, Germany

a r t i c l e i n f o a b s t r a c t

Article history: Objective: To assess the impact of epilepsy and antiseizure medications (ASMs) on sleep quality in people
Received 29 June 2023 with epilepsy (PWE).
Revised 29 September 2023 Methods: An online survey was conducted in France, Germany, Italy, Spain and the UK among PWE taking
Accepted 29 September 2023
>1 ASM and matched controls. Sleep quality was evaluated using the Pittsburgh Sleep Quality Index
Available online 19 October 2023
(PSQI). Associations between sleep quality (global PSQI) and overall quality of life (QoL; assessed using
the 12-Item Short Form Survey [SF-12]) and sleep quality and depressive symptoms (assessed using
Keywords:
the Neurological Disorders Depression Inventory for Epilepsy [NDDI-E]) were also evaluated.
Antiseizure medication
Epilepsy
Results: Overall, 500 PWE and 500 matched controls were included. PWE had significantly greater mean
Quality of life global PSQI scores than controls (9.32 vs 7.56; p < 0.0001), with 80% reporting a score >5 versus 66% of
Sleep controls (p < 0.001). PWE experienced significantly more problems with most PSQI components than con-
People with epilepsy trols. Mean global PSQI scores in PWE receiving 2 versus
3 ASMs were 9.03 and 10.18, respectively (p <
0.004); global PSQI scores >5 were reported in 76% versus 90%, respectively (p = 0.001). Regimens con-
taining lamotrigine or phenobarbital were associated with poorer sleep quality than those without these
ASMs. In PWE, negative correlations were identified between global PSQI scores and both the SF-12
physical and mental components (Pearson’s correlation coefficient [PCC], 0.61 and 0.40, respectively);
NDDI-E and global PSQI scores were positively correlated (PCC, 0.6).
Conclusions: PWE experience significantly worse sleep quality than people without epilepsy, with some
ASMs contributing to poorer sleep. QoL and physical and mental health were all affected by sleep quality
in PWE.
Ó 2023 The Author(s). Published by Elsevier Inc. This is an open access article under the CC BY license
(http://creativecommons.org/licenses/by/4.0/).

1. Introduction
Abbreviations: ADM, Arbeitskreis Deutscher Markt- und Sozialforschungsinsti-
tute; ASM, antiseizure medication; BVM, Berufsverband Deutscher Markt- und Sleep is an essential process for an individual’s health and well-
Sozialforscher; BRV, brivaracetam; CBZ, carbamazepine; LTG, lamotrigine; LCM, being, as it plays a major role in several brain functions, such as
lacosamide; LEV, levetiracetam; MDD, major depressive disorder; NDDI-E, Neuro- cognitive and safety-related performance, memory consolidation,
logical Disorders Depression Inventory for Epilepsy; OXC, oxcarbazepine; PER, and mood regulation, as well as in regular systemic physiology
perampanel; PB, phenobarbital; PSQI, Pittsburgh Sleep Quality Index; PWE, people
with epilepsy; QoL, quality of life; SF-12, 12-Item Short Form Survey; SD, standard
[1,2].
deviation; VPA, valproate; ZNS, zonisamide. Sleep disturbances have a deleterious effect on people’s health,
⇑ Corresponding author at: Epilepsy Center Frankfurt Rhine-Main, Department of having both short- and long-term consequences. In the short term,
Neurology, Goethe-University, Frankfurt Schleusenweg 2-16, 60528 Frankfurt am sleep disturbances have been associated with stress responsivity,
Main, Germany.
somatic pain, reduced quality of life (QoL), emotional distress
E-mail address: strzelczyk@med.uni-frankfurt.de (A. Strzelczyk).

https://doi.org/10.1016/j.yebeh.2023.109481
1525-5050/Ó 2023 The Author(s). Published by Elsevier Inc.
This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/).
C. Lawthom, A. Didelot, A. Coppola et al. Epilepsy & Behavior 148 (2023) 109481

and mood disorders, and cognitive, memory, and performance def- 59 years and had a formal diagnosis of epilepsy with at least one
icits [3]. The long-term consequences of sleep disruption include seizure in the past 12 months; PWE had to be treated with more
increased risks of hypertension, dyslipidemia, cardiovascular dis- than one ASM, but a maximum of 30% of the total PWE population
ease, weight-related issues, metabolic syndrome, type 2 diabetes included could be on
3 ASMs. The control group was recruited
mellitus, and colorectal cancer [2,3]. after PWE so that controls were matched in terms of age, gender,
Sleep disturbances are twice as common in people with epi- level of education, and employment status.
lepsy (PWE) compared with healthy individuals [4], and it is esti-
mated that approximately one-third of PWE report at least one 2.2. Study outcomes
sleep disorder [4]. Sleep disturbances affecting PWE include
insomnia, excessive daytime sleepiness, sleep apnea, and restless 2.2.1. Pittsburgh Sleep Quality Index
leg syndrome [4,5]. The quality of sleep was evaluated using the Pittsburgh Sleep
The relationship between sleep and epilepsy is complex and Quality Index (PSQI) [21]. This is a self-report questionnaire that
bidirectional. On the one hand, interictal epileptiform discharges assesses sleep quality over a 1-month time interval across seven
and seizures can be activated during the non-rapid eye movement sleep components. The PSQI consists of 19 items that are combined
(NREM) sleep phase [6,7]. On the other hand, seizures can disrupt to produce a sub-score for each component, with each sub-score
sleep architecture by reducing and interfering with the rapid-eye- ranging from 0 to 3 whereby ‘00 indicates no sleep difficulties
movement (REM) sleep phase, decreasing total sleep time, lower- and ‘30 represents more severe sleep problems. The global PSQI
ing sleep efficiency, and increasing sleep fragmentation, sleep score is the sum of the sub-scores of the seven components and
latency, sleep stage shifts, and awakenings [8]. Additionally, sleep ranges from 0 to 21, in which a higher score indicates worse sleep
deprivation is a trigger for seizures, and this can lead to a negative quality and a total score above five indicates poor sleep quality.
vicious cycle for PWE [7,9]. Furthermore, evidence suggests a The seven components are: sleep duration (how many hours of
decrease in sleep quality and increased levels of anxiety and sleep a person actually gets at night); sleep disturbance (derived
depression among caregivers of PWE relative to the general popu- from the sum of eight item scores indicating how often the person
lation [10]. had trouble sleeping); sleep latency (how long it takes to go to
It is well known that antiseizure medications (ASMs) can also sleep); daytime dysfunction (due to sleepiness); sleep efficiency
affect sleep quality and sleep architecture: some ASMs have been (calculated as a ratio of number of hours slept/number of hours
shown to increase NREM and reduce REM; other ASMs reduce spent in bed multiplied by 100, which translates into a percentage
sleep latency and efficiency; and a number of ASMs can also cause (>85% indicates no sleep problems, scored as 0; <65% indicates
daytime sleepiness or insomnia [9,11,12]. ASMs can also have an more severe sleep problems, scored as 3), overall sleep quality
indirect impact on sleep. Several ASMs are associated with weight (person’s rating of their overall sleep quality on a scale of 0 (very
gain, which can independently increase the risk of conditions such good) to 3 (very bad); and sleep medication use. [21,22]. In the cur-
as obstructive sleep apnea [11]. rent study, global PSQI scores >5 were further divided into two cat-
A European survey of 500 PWE matched to 500 controls was egories, 6–11 and >11, to provide more granularity around the PSQI
conducted to assess how epilepsy and its treatment affect and scores and to differentiate between ‘poor sleepers’ and ‘very poor
interfere with the lives of PWE [13]. The survey reported that, com- sleepers’. The full PSQI was used (as opposed to a shortened ver-
pared with controls, PWE have a poorer QoL and are significantly sion) in order to provide an in-depth assessment of sleep and to
more likely to suffer from depressive symptoms [13], in line with allow participants to be categorized into good and poor sleepers
previous findings [14–19]. The survey highlighted that epilepsy [21]. The PSQI scores were also compared between the subgroups
has a negative impact on the QoL and on the physical and mental of PWE receiving different treatment regimens (containing the 10
health of PWE, interfering with their daily activities, studies or most commonly used ASMs); treatment regimens containing one
work, and social life [13]. It also suggested that epilepsy treatment of the following ASMs, brivaracetam (BRV), carbamazepine (CBZ),
has an impact on PWE’s QoL, as PWE on
3 ASMs suffered from lacosamide (LCM), lamotrigine (LTG), levetiracetam (LEV), oxcar-
poorer QoL than PWE on only two ASMs [13]. Presented here are bazepine (OXC), perampanel (PER), phenobarbital (PB), valproate
the results of further assessments from the European survey, (VPA), or zonisamide (ZNS), were each compared with all the other
which addressed how epilepsy affects the sleep quality and habits treatment regimens.
of PWE. Correlations between the global PSQI score and scores from the
12-item Short Form Survey (SF-12) [23] and the Neurological
Disorders Depression Inventory for Epilepsy (NDDI-E) [24] were
2. Methods additionally evaluated. The SF-12 is a self-reported questionnaire
utilized to assess the impact of health on individuals’ QoL. It con-
2.1. Study design and patient population sists of 12 questions, creating eight domains with results calcu-
lated via a scoring program [23]. Scores <50 for the physical
This was a 30-minute online survey conducted in five European component and <42 for the mental component were considered
countries: France, Germany, Italy, Spain, and the UK. The details of below the norm [25]. The NDDI-E is a self-reported questionnaire
the study design have been previously published [13]. All partici- for screening major depressive disorder (MDD) in PWE, with scores
pants provided informed consent to participate in the study, and summed to calculate an overall score indicating the likelihood of
the research was carried out in line with relevant codes of conduct MDD; overall scores above 15 were considered positive for depres-
and guidance for market research, including the recommendations sive symptoms [24].
of the European Society for Opinion and Marketing Research, the Respondents were also asked questions about their sleep habits
European Pharmaceutical Market Research Association, the British and arrangements.
Healthcare Business Intelligence Association, and (in Germany
only) the Arbeitskreis Deutscher Markt- und Sozialforschungsinsti- 2.3. Statistical analyses
tute[20]/Berufsverband Deutscher Markt- und Sozialforscher
(BVM). The data were analyzed using SPSS and Q. Statistical differences
The inclusion and exclusion criteria have been previously for continuous variables were assessed using the student’s t-test. A
described [13]. Briefly, PWE were included if they were aged 18– p-value of <0.05 was considered significant for single comparisons.
2
C. Lawthom, A. Didelot, A. Coppola et al. Epilepsy & Behavior 148 (2023) 109481

A Bonferroni correction was applied for data with multiple com- Table 1
parisons (e.g., mean PSQI sub-scores for PWE and controls). In PSQI sub-scores in PWE and controls.

these cases, a p-value of <0.05 / (number of comparisons) was con- PWE mean Controls mean p-value
sidered significant. Statistical differences for proportions of cate- score score
gorical rating data were assessed using a Z test for proportions. Subjective sleep 1.36 1.38 0.73
Correlations between the SF-12, NDDI-E, and PSQI scores were quality
assessed using the Pearson correlation. A chi-square test of inde- Sleep latency 1.65 1.48 0.003
Sleep duration 0.90 1.00 0.07
pendence was performed to assess the significance of the PSQI Sleep efficiency 1.14 0.73 < 0.0001
scores between PWE and the controls. Sleep disturbance 1.56 1.33 < 0.0001
Sleep medication 1.24 0.55 < 0.0001
Daytime dysfunction 1.47 1.09 < 0.0001
3. Results
Significant values are reported in bold. Statistical relevance tested via a t-test, and
as the analysis was run on multiple sub-sets of the PSQI, a Bonferroni correction
3.1. Patient population was applied.
PSQI, Pittsburgh Sleep Quality Index; PWE, people with epilepsy.
In total, 500 PWE were matched to 500 controls (100 PWE and
100 matched controls from each of the five countries) [13]. The
baseline and disease characteristics have been previously 3.2.1. Subjective sleep quality
described: in both groups, 47% of respondents were women, and No significant differences were identified in the PSQI sub-scores
most participants (approximately 70%) were aged 30–49 years. for subjective sleep quality between PWE and controls (Fig. 2).
Both PWE and controls reported a number of comorbidities, Similar proportions of PWE and controls reported scores of 0, 1,
including migraine, mood disorders, osteoporosis, and chronic dis- 2, and 3.
eases such as diabetes [13]. The overall sleep quality was rated as ‘very good’ by 12% of PWE
The mean time since epilepsy diagnosis was 18.2 years, and and 9% of controls and as ‘fairly good’ by 46% of PWE and 50% of
most PWE (65%) identified themselves as having focal-onset epi- controls; 37% of PWE and 35% of controls rated it as ‘fairly bad’
lepsy. The median number of seizures per year was 5, and the med- and 5% of PWE and 6% of controls rated it as ‘very bad’.
ian time since the last seizure occurrence was 3 months. Overall, PWE receiving two ASMs reported better quality of sleep than
75% of PWE were taking two ASMs, and 25% were taking
3 ASMs. those receiving
3 ASMs. Significantly more PWE on two ASMs
reported ‘very good’ sleep quality compared with those on
3
ASMs (13% vs. 6%; p = 0.03); 51% of those receiving
3 ASMs rated
3.2. Pittsburgh Sleep Quality Index
their sleep quality as ‘fairly to very bad’ compared with 40% of
those on two ASMs (p = 0.03).
The mean global PSQI score was significantly higher in PWE
than in controls (9.32 vs. 7.56; p < 0.0001), with 80% of PWE report-
ing a score >5 compared with 66% of controls (p < 0.001) (Fig. 1), in 3.2.2. Sleep latency
which higher scores indicate poorer sleep quality. PWE receiving PWE were significantly less likely to have a sub-score of 0 for
two ASMs had a significantly lower mean global PSQI score than sleep latency than controls (8% vs. 15%; p < 0.001), and they were
PWE on
3 ASMs (9.03 vs. 10.18; p < 0.004) and were significantly significantly more likely to have a sub-score of 2 (45% vs. 30%;
less likely to report a global PSQI score of >5 (76% vs. 90%; p < 0.001) (Fig. 2). It took PWE three times longer to fall asleep ver-
p = 0.001; Fig. 1), indicating that PWE receiving a lower number sus controls (99 vs. 35 min; p < 0.0001).
of ASMs have better sleep quality. The chi-square test (33.054;
degree of freedom [df] = 2; p < 0.001) indicated a significant asso- 3.2.3. Sleep duration and habitual sleep efficiency
ciation between epilepsy and poor sleep. Overall, significantly more PWE than controls scored 0 for sleep
Compared with controls, PWE experienced significantly more duration (37% vs. 26%; p < 0.001) (Fig. 2). Significantly higher pro-
problems with most of the seven PSQI components, except for sub- portions of PWE than controls had a sub-score of 3 (17% vs. 10%;
jective sleep quality and sleep duration (Table 1; Fig. 2). p < 0.001) and 2 (20% vs. 10%; p < 0.0001) for sleep efficiency

Fig. 1. Global PSQI scores in all PWE, in PWE receiving two ASMs and
3 ASMs, and controls. A score >5 is indicative of poor sleep quality, and a higher total score
indicates worse sleep quality. ASM, antiseizure medication; PSQI, Pittsburgh Sleep Quality Index; PWE, people with epilepsy.

3
C. Lawthom, A. Didelot, A. Coppola et al. Epilepsy & Behavior 148 (2023) 109481

Fig. 2. Mean PSQI scores for each of the seven PSQI components in PWE and controls. p-values are for PWE vs. controls. PSQI, Pittsburgh Sleep Quality Index; PWE, people
with epilepsy.

(Fig. 2). PWE reported going to bed earlier than controls (11 pm vs. 27% less than once a week, compared with 8%, 10%, and 12% of con-
midnight) with a similar wake-up time of 8am, but the amount of trols, respectively (p < 0.001 for all comparisons), with no medica-
sleep per night was similar between the two groups (PWE: 6.8 h; tions being taken by 32% of PWE and 70% of controls (p < 0.0001).
controls: 6.7 h). PWE who had problems with sleep duration were more likely than
those without to have taken medicine to help them sleep once or
3.2.4. Sleep disturbances twice a week (29% vs. 19%, p = 0.0107) or three or more times
Significantly more PWE than controls had a sub-score of 3 (7% per week (21% vs. 6%, p = 0.00002).
vs. 3%; p = 0.001) and 2 (47% vs. 33%; p < 0.0001) for sleep distur-
bances (Fig. 2). Compared with controls, PWE had significantly
more trouble sleeping due to pain, bad dreams, feeling too hot or 3.2.6. Daytime dysfunction
too cold, coughing or snoring loudly, and being unable to breathe More PWE than controls had a sub-score of 3 (10% vs. 4%;
easily (Fig. 3). p < 0.001) and 2 (39% vs. 26%; p < 0.0001) for daytime dysfunction
PWE with a poorer sleep quality (‘very bad’ or ‘fairly bad’ qual- (Fig. 2). PWE were significantly more likely than controls to have
ity) more frequently had trouble sleeping due to pain than those trouble staying awake, carrying out tasks requiring good attention
with a good sleep quality (‘very good’ or ‘fairly good’ quality). Pain or alertness, or taking part in social activity. During the past month,
impacted sleep at least three times a week in 17% and 5% of PWE 9% of PWE and 4% of controls reported issues
3 times a week
with poor and good sleep quality, respectively (p < 0.0001). (p = 0.01); 30% of PWE and 14% of controls reported issues once
Whereas only 18% of PWE with poorer sleep quality did not report or twice a week (p < 0.0001); 30% of PWE and 21% of controls
trouble sleeping due to pain in the past month, the proportion of reported issues less than once a week (p = 0.002); and 32% of
PWE with good sleep quality without this issue was 51% PWE and 61% of controls had no issues (p < 0.0001).
(p < 0.0001). PWE were significantly more likely than controls to have ‘quite
a problem’ (34% vs. 26%, p = 0.007) with keeping up enough enthu-
3.2.5. Use of sleeping medications siasm to get things done, and were less likely to not report any
A significantly higher proportion of PWE scored 3 (15% vs. 8%; problem at all (18% vs. 29%, p = 0.00004). PWE on
3 ASMs were
p < 0.001) or 2 (25% vs. 10%; p < 0.0001) for use of sleep medication significantly more likely than those on two ASMs to have ‘quite a
compared with controls (Fig. 2). In the last month, 15% of PWE took problem’ (43% vs. 31%, p = 0.02) and ‘a very big problem’ (14% vs.
sleep medications
3 times a week, 25% once or twice a week, and 7%, p = 0.02).

Fig. 3. Reasons for having trouble sleeping in PWE and controls. p-values are for PWE vs. controls; a chi-square test of independence was used to assess the significance
between PWE vs. controls. PWE, people with epilepsy.

4
C. Lawthom, A. Didelot, A. Coppola et al. Epilepsy & Behavior 148 (2023) 109481

3.2.7. Pittsburgh Sleep Quality Index scores in relation to different components were significantly lower in PWE with a PSQI global
treatment regimens score
11 compared with PWE with a PSQI global score of 0–5
The global PSQI score was significantly higher in the subgroup and with those with a score of 6–10 (except for vitality) (Supple-
of PWE receiving a treatment regimen containing LTG than in mentary Fig. S1b).
PWE receiving treatment regimens without LTG (10.19 vs. 8.89;
p = 0.0006) and in the subgroups of patients receiving a treatment 3.4. Correlation between the Pittsburgh Sleep Quality Index and
regimen containing PB than in PWE on regimens without PB (10.54 mental health and mood in people with epilepsy
vs. 9.17; p = 0.0275). The global PSQI scores were not statistically
different for the treatment regimens containing BRV, CBZ, LCM, A moderately positive correlation between the PSQI score and
LEV, OXC, PER, VPA or ZNS, compared with other treatment regi- NDDI-E score among PWE was identified (Pearson correlation coef-
mens (Fig. 4). ficient, 0.6) (Fig. 7). The mean NDDI-E score was significantly
There were some noticeable differences in the total sub-scores higher in PWE with a global PSQI score
11 (17.4) than in PWE
of each PSQI component between the different treatment regimens with a global PSQI score of 6–10 (13.6; p < 0.001) and in PWE with
(Fig. 5). PWE receiving LTG were significantly more likely to have a global PSQI score of 0–5 (10.0; p < 0.001). PWE with a PSQI global
higher sub-scores for sleep disturbances, use of sleep medications, score
11 were significantly more likely to have an NDDI-E score
and daytime dysfunction than PWE on other treatment regimens of 15–24 (77%) than PWE with a PSQI global score of 0–5 (21%)
(Supplementary Table S1). PWE receiving PB were significantly and of 6–10 (47%) (p < 0.0001 for both comparisons) (Supplemen-
more likely to have a higher sub-score for sleep duration and sleep tary Fig. S2).
disturbances than PWE on other regimens. Treatment regimens
containing BRV, OXC, or PER were significantly more likely to be 3.5. Additional questions on sleeping arrangements and habits
associated with lower sub-scores for subjective sleep quality than
PWE on other treatment regimens. PWE receiving PER were also 3.5.1. Sleep habits
significantly more likely to score 0 for sleep duration and daytime Controls were significantly more likely than PWE to share a bed
dysfunction than PWE on other treatment regimens (Supplemen- with someone (54% vs. 31%; p < 0.0001), although PWE were signif-
tary Table S1). Treatment with CBZ was associated with signifi- icantly more likely than controls to have someone in another room
cantly lower sub-score for sleep duration than other regimens, (25% vs. 18%; p = 0.005) or in the same room but not the same bed
with PWE on CBZ significantly more likely to score 0 (i.e., claiming (21% vs. 4%; p < 0.0001).
that they are sleeping more than 7 h per night) than PWE on other PWE were significantly more likely than controls to have expe-
treatment regimens. VPA was associated with significantly higher rienced long pauses between breaths while asleep (13% vs. 3%;
sub-scores for subjective sleep quality and daytime dysfunction. p < 0.0001), legs twitching or jerking while asleep (21% vs. 9%;
No significant differences were identified in the PSQI sub-scores p < 0.0001), periods of disorientation or confusion when waking
for treatment regimens containing LCM, LEV, or ZNS compared up (12% vs. 5%; p = 0.002) at night at least three times a week. They
with other treatment regimens. were also more likely to experience periods of loud snoring (27%
vs. 18%; p = 0.002) and other types of restlessness (28% vs. 13%;
3.3. Correlation between the Pittsburgh Sleep Quality Index and p = 0.003) while they were asleep once or twice a week.
quality of life in people with epilepsy
3.5.2. Impact of epilepsy on sleep and sleep habits
Among PWE, a moderate negative correlation was identified Control patients did not spontaneously mention problems with
between the physical component of the SF-12 and the global PSQI sleep quality, ability to fall asleep, or staying asleep, whereas PWE
score (Pearson correlation coefficient, 0.61), and a low negative reported several issues, and the top spontaneous mentions of the
correlation was identified between the mental component and effects of epilepsy on sleep included poor sleep quality (mentioned
the PSQI total score (Pearson correlation coefficient, 0.40) by 18% of PWE), panic/anxiety (13%), difficulty falling asleep (10%),
(Fig. 6). Over half of PWE with a global PSQI score
11 had an difficulty staying asleep (10%), feeling badly affected (8%), agita-
SF-12 score <40 for the physical (52%) and the mental component tion/restlessness (6%), fatigue (5%), affected family life (4%), and
(58%) (Supplementary Fig. S1a). Average SF-12 scores for all eight sleep only improved by medications (4%). Factors that impact sleep

Fig. 4. Global PSQI scores in PWE receiving a treatment regimen containing one of the analyzed ASMs (BRV, CBZ, LCM, LTG, LEV, OXC, PER, PB, VPA, or ZNS) vs. other
treatment regimens. Statistically significant values are bolded. ASM, antiseizure medication; BRV, brivaracetam; CBZ, carbamazepine; LTG, lamotrigine; LCM, lacosamide;
LEV, levetiracetam; NS, not significant; OXC, oxcarbazepine; PER, perampanel; PB, phenobarbital; PSQI, Pittsburgh Sleep Quality Index; PWE, people with epilepsy; SD,
standard deviation; VPA, valproate; ZNS, zonisamide.

5
C. Lawthom, A. Didelot, A. Coppola et al. Epilepsy & Behavior 148 (2023) 109481

Fig. 5. Sub-scores for each PSQI component in PWE receiving a treatment regimen containing one of the analyzed ASMs (BRV, CBZ, LCM, LTG, LEV, OXC, PER, PB, VPA,
or ZNS) vs. other treatment regimens. Statistically significant values are bolded. ASM, antiseizure medication; BRV, brivaracetam; CBZ, carbamazepine; LTG, lamotrigine;
LCM, lacosamide; LEV, levetiracetam; NS, not significant; OXC, oxcarbazepine; PER, perampanel; PB, phenobarbital; PSQI, Pittsburgh Sleep Quality Index; PWE, people with
epilepsy; SD, standard deviation; VPA, valproate; ZNS, zonisamide.

in the control patients were more often related to external factors 4. Discussion
than their ability to sleep (general stress [14%], noise disturbance
[10%], work changes [7%], weather or temperature [6%], children This study reports the results of a survey that assessed the
disturbances [5%], multiple thoughts [5%], and economic problems impact of epilepsy and its treatment on sleep.
[4%].

6
C. Lawthom, A. Didelot, A. Coppola et al. Epilepsy & Behavior 148 (2023) 109481

Fig. 6. Correlation between the PSQI global score and the physical and mental components of the SF-12 in PWE. PSQI, Pittsburgh Sleep Quality Index; PWE, people with
epilepsy; SF-12, 12-Item Short Form Survey.

Fig. 7. Correlation between the NNDI-E score and the PSQI in PWE. NDDI-E, Neurological Disorders Depression Inventory for Epilepsy; PSQI, Pittsburgh Sleep Quality
Index; PWE, people with epilepsy.

Overall, a significantly higher proportion of PWE reported hav- important to remember that the PSQI only measures sleep quality
ing poor sleep quality (PSQI global score >5) than controls (80% vs. in the past month.
66%; p < 0.001). Compared with controls, PWE reported signifi- The association between epilepsy and poor sleep quality has
cantly worse PSQI scores for sleep latency, efficiency, disturbance, also been reported in other studies. In a questionnaire involving
use of sleep medication, and daytime dysfunction. 208 PWE and 212 controls, 30% of PWE reported bad sleep quality
PWE with trouble sleeping reported bad dreams, pain, inability (global PSQI score 6–21), compared with only 16% of controls. In
to breathe easily, feeling cold and coughing, or snoring loudly as particular, PWE were more likely to complain about symptoms of
key reasons for their trouble sleeping; in particular, pain seems to insomnia, sleep apnea, and parasomnia [28].
be a key factor interfering with PWE’s sleep quality. PWE also A National Health Service study conducted in the US compared
complained about loud snoring, long pauses between breaths, sleep quality in individuals without epilepsy (N = 93,126) with
twitching legs, disorientation, and other restlessness more fre- PWE with any epilepsy (N = 1,774), PWE with active epilepsy
quently in the past month than the controls. These symptoms (PWE taking ASMs and/or experiencing at least one seizure in the
were significantly more common in PWE than in controls and past year; N = 1,101), and PWE with inactive epilepsy (PWE with
might be related to the presence of comorbidities and/or sleep a diagnosis of epilepsy but not taking ASMs and not having expe-
disorders, such as sleep apnea, which have been shown to be rienced a seizure in the past year; N = 673) [29]. Compared with
more common in PWE than in the general population [4,26]. the controls, PWE in all three groups had a higher prevalence of
For this reason, PWE would have been expected to report a lower short sleep duration (<7 h of sleep per night), long sleep duration
subjective quality of sleep than controls, but the score in relation (>9 h per night), and a lower prevalence of healthy sleep duration.
to this component was similar between the two groups. It should Moreover, compared with the controls, PWE in all categories had a
be noted that the PSQI measures ‘subjective’ sleep quality. The higher prevalence of trouble falling asleep (
3 times in the past
perception of a ‘good night’s sleep’ might therefore be different week), trouble staying asleep (
3 times in the past week), use of
for each individual, and it might also be different for control medication to fall asleep or stay asleep (
1 time in the past week),
patients in comparison with people living with uncontrolled epi- and non-restorative sleep (
3 d in the past week) [29]. Similarly, a
lepsy, who might be afraid of suffering seizures during the night. large meta-analysis that included 2,964 PWE and 5,232 controls
Controls recruited in this survey were not ‘healthy’ but reported a found that PWE had a significantly higher global PSQI score than
number of comorbidities, including long-term conditions, which controls (p < 0.001) [30], and in a cross-sectional study of 152
could have affected their sleep quality [27]. Finally, it is also PWE and 152 controls, daytime sleepiness, difficulty in sleep main-

7
C. Lawthom, A. Didelot, A. Coppola et al. Epilepsy & Behavior 148 (2023) 109481

tenance, poor sleep quality, and restless leg syndrome were found epilepsy on sleep, which included poor sleep quality, panic or anx-
to be more common in PWE than in controls [31]. In a meta- iety, difficulty falling asleep, difficulty staying asleep, feeling badly
analysis of 16 studies, 12 used the PSQI to evaluate sleep quality, affected, and agitation or restlessness. PWE complained about how
and in eight of these studies, significantly higher global PSQI scores epilepsy affects sleep, the emotional and psychological impact of
were reported for PWE compared with controls [32]. epilepsy, and how these factors interfere with their daily lives. Epi-
Our study identified differences in the sleep quality of PWE in lepsy impacts sleep and sleeping habits in a number of negative
relation to different treatment regimens. LTG and PB were both ways, not only resulting in difficulties sleeping, but also in emo-
associated with significantly worse overall sleep quality than other tional and psychological distress and problems during waking life
ASMs, as suggested by the global PSQI scores. Although the global (e.g., fatigue, impact on family life). PWE’s sleeping arrangements
PSQI scores indicate that there were no significant differences in also indicated that PWE feel anxious about sleep: PWE are more
the overall sleep quality between the other ASMs, the sub-scores likely than controls to have someone in the room or in the house,
of each PSQI component confirm that ASMs can have a significant probably reflecting PWE’s fear and anxiety of experiencing seizures
impact (positive or negative) on different aspects of sleep. Indeed, at night. PWE’s complaints about not being able to take part in
our findings suggest that PER, OXC, and BRV might have a positive social activities and not being able to keep up enough enthusiasm
effect on subjective sleep quality, with PER also having a positive to get things done might also be related to their sleep concerns and
effect on sleep duration and daytime dysfunction; CBZ might poor sleep quality.
improve sleep duration; and VPA might have a negative effect on The results are in line with the findings of a qualitative netno-
both subjective sleep quality and daytime dysfunction. A recent graphic study of epilepsy conversations posted on public social
review analyzed available data on the effect of ASMs on sleep media sites, which highlighted that PWE are concerned about
architecture [11]. The study concluded that eslicarbazepine acet- sleep, about its connection with seizures, and about the necessity
ate, LCM, and PER improved or had no effect on sleep; PER was for strict sleep schedules. It also revealed that PWE tend to worry
associated with a low incidence of insomnia and LCM with a low excessively about the need to achieve good-quality sleep, poten-
incidence of daytime sleepiness. Clonazepam, felbamate, LTG, tially increasing their anxiety and worsening pre-existing mental
OXC, and PB worsened or had no effect on sleep, with LTG possibly health issues [46]. On social media, PWE also complained that their
associated with insomnia and PB with daytime sleepiness. strict sleep schedules have a deleterious impact on their social life
Cannabidiol, CBZ, and LEV had no effect on sleep, and mixed data and work [46].
on VPA were collected. Our results are broadly in line with the As previously mentioned [13], the current study has several
findings of the review, indicating that ASMs affect sleep in different limitations. It is possible that a selection bias in favor of a younger,
ways and that this factor should be considered when selecting an more urban, and educated population has been introduced as this
ASM. was an online survey. The voluntary participation in the survey
The current study identified links between QoL and sleep qual- might have also introduced bias into the type of individuals it
ity and between depression and sleep quality. PWE with poorer included. It should be taken into account that only European coun-
sleep quality were significantly more likely to have lower SF-12 tries were included; therefore, the sample of PWE included in this
scores (indicative of poor QoL) as well as higher NDDI-E scores study represents a specific sociocultural and economic region. Fur-
(indicative of depressive symptoms). These findings support previ- thermore, PWE included in the study had uncontrolled epilepsy,
ous analyses of the association between QoL, mental health, and suffering from at least one seizure a year, and represented a
sleep quality. Associations between depression and mood disor- heterogenous group of PWE but did not necessarily represent the
ders, sleep disturbances, and poor QoL have been reported in both most severe cases of epilepsy. Another important point to note is
individuals without epilepsy and PWE [33–37]. In a study of 124 that the survey was conducted during the COVID-19 pandemic
PWE conducted in Greece, the presence of excessive daytime [47], which could have impacted the mood and the mental and
sleepiness, obstructive sleep apnea, and insomnia was associated physical health of participants and might have affected sleep qual-
with impaired QoL, with insomnia particularly affecting emotional ity. It should also be considered that the PSQI used in the survey is
well-being, energy/fatigue, cognitive functioning, and social func- a subjective tool to evaluate sleep quality and that the results were
tioning[36]. In a cross-sectional observational study involving 84 not confirmed with the use of objective sleep quality measure-
PWE, poor QoL was found to be associated with high levels of anx- ments. Finally, PWE were not receiving ASM monotherapy but
iety and depression and also with sleep disturbance, waking up treatment regimens containing 2–
3 ASMs; therefore, findings
during the night, and problems falling asleep, especially in individ- on the effect of ASMs on sleep might have also been related to
uals with drug-resistant epilepsy (DRE) [35]. Similarly, the current the interactions between different agents. Future studies should
study reported that PWE on
3 ASMs (generally accepted as being look at the causal relationship between sleep, seizures, and mood
indicative of DRE [38,39]) suffered from poorer sleep quality com- in PWE.
pared with those on only two ASMs, with this association poten-
tially related to adverse events associated with polytherapy [40–
42], which might also have an impact on sleep. 5. Conclusions
The results of the current survey suggest that PWE are deeply
concerned about their sleep: PWE considered achieving good In summary, this study reported that epilepsy has a significant
sleeping habits a priority and mentioned sleep as one of the key impact on PWE’s sleep, influencing PWE’s overall QoL and physical
factors affected by epilepsy [13], confirming previous reports that and mental health, and that the use of ASMs also affects sleep qual-
PWE complain about sleep issues [9,43–45]. The early bedtime ity. It also suggests that PWE are aware of the importance of sleep
and frequent use of sleep medication in PWE might be related to but might worry excessively about their sleep habits, which might
their awareness of the importance of achieving good-quality sleep; represent a further risk for seizure recurrence.
however, the fact that PWE need a long time to fall asleep (over
90 min) suggests that they are anxious and worried about their
sleep habits, which is likely to affect their ability to fall asleep. This Data statement
might create a vicious cycle, further aggravating sleep disturbances
with the risk of increasing seizure frequency. This hypothesis is The data that supports the findings of this study are available
also supported by the top spontaneous mentions of the effects of from the authors upon reasonable request.
8
C. Lawthom, A. Didelot, A. Coppola et al. Epilepsy & Behavior 148 (2023) 109481

Declaration of competing interest [15] Izci F, Fındıklı E, Camkurt MA, Tuncel D, Sß ahin M. Impact of aggression,
depression, and anxiety levels on quality of life in epilepsy patients.
Neuropsychiatr Dis Treat 2016;12:2595–603.
The authors declare the following financial interests/personal [16] Siarava E, Hyphantis T, Katsanos AH, Pelidou SH, Kyritsis AP, Markoula S.
relationships which may be considered as potential competing Depression and quality of life in patients with epilepsy in Northwest Greece.
Seizure 2019;66:93–8.
interests: CL has received speaker fees from UCB Pharma and Eisai,
[17] Mula M, Sander JW. Psychosocial aspects of epilepsy: a wider approach.
advisory board fees from GW Pharma, and contributed to manu- BJPsych Open 2016;2(4):270–4.
script productions for Healios. CL is employed by Lundbeck in Med- [18] Keezer MR, Sisodiya SM, Sander JW. Comorbidities of epilepsy: current
concepts and future perspectives. Lancet Neurol 2016;15(1):106–15.
ical Affairs in a non-epilepsy-related role. AD has received speaker
[19] Jacoby A, Sudell M, Tudur Smith C, Crossley J, Marson AG, Baker GA, et al.
fees from EISAI and consultancy fees from Angelina Pharma, EISAI, Quality-of-life outcomes of initiating treatment with standard and newer
TEVA Sante, and UCB. AC has received speaker fees from EISAI antiepileptic drugs in adults with new-onset epilepsy: findings from the
and GW/Jazz pharmaceutical company and consultancy fees by SANAD trial. Epilepsia 2015;56(3):460–72.
[20] U.S. Food and Drug Administration. Ongentys Prescribing Information, 2020.
GW/Jazz pharmaceutical company, UCB, and BIAL. AAS has partici- https://www.accessdata.fda.gov/drugsatfda_docs/label/2020/212489s000lbl.
pated to educational and research activities funded by GW, Bial, pdf. Accessed 13 October 2023.
Eisai, UCB, Zogenix, PTC, and Angelini. BF is an employee of Eisai. [21] Buysse DJ, Reynolds 3rd CF, Monk TH, Berman SR, Kupfer DJ. The Pittsburgh
Sleep Quality Index: a new instrument for psychiatric practice and research.
RSF is an employee of Eisai. AS has received personal fees and Psychiatry Res 1989;28(2):193–213.
grants from Angelini Pharma/Arvelle Therapeutics, Desitin [22] Cox SD, Benoit JS, Brohard CL, McIntyre TM. Evaluation of sleep quality among
Arzneimittel, Eisai, Jazz/GW Pharmaceuticals, Marinus Pharma, nursing faculty: application of the Pittsburgh Sleep Quality Index-A
descriptive correlational study. Nurs Open 2022;9(1):339–48.
Medtronic, Takeda, UCB, UNEEG Medical, and Zogenix. [23] Huo T, Guo Y, Shenkman E, Muller K. Assessing the reliability of the short form
12 (SF-12) health survey in adults with mental health conditions: a report
from the wellness incentive and navigation (WIN) study. Health Qual Life
Acknowledgments Outcomes 2018;16(1):34.
[24] Friedman DE, Kung DH, Laowattana S, Kass JS, Hrachovy RA, Levin HS.
This study was funded by Eisai Ltd. Editorial assistance was pro- Identifying depression in epilepsy in a busy clinical setting is enhanced with
systematic screening. Seizure 2009;18(6):429–33.
vided by Eliana D’Araio of mXm Medical Communications and [25] Ware Jr J, Kosinski M, Keller SD. A 12-Item Short-Form Health Survey:
funded by Eisai Ltd. construction of scales and preliminary tests of reliability and validity. Med
Care 1996;34(3):220–33.
[26] Gutter T, Callenbach PMC, Brouwer OF, de Weerd AW. Prevalence of sleep
Appendix A. Supplementary data disturbances in people with epilepsy and the impact on quality of life: a survey
in secondary care. Seizure 2019;69:298–303.
[27] Kemple M, O’Toole S, O’Toole C. Sleep quality in patients with chronic illness. J
Supplementary data to this article can be found online at Clin Nurs 2016;25(21–22):3363–72.
https://doi.org/10.1016/j.yebeh.2023.109481. [28] Ismayilova V, Demir AU, Tezer FI. Subjective sleep disturbance in epilepsy
patients at an outpatient clinic: a questionnaire-based study on prevalence.
Epilepsy Res 2015;115:119–25.
References [29] Tian N, Wheaton AG, Zack M, Croft JB, Greenlund KJ. Sleep duration and quality
among U.S. adults with epilepsy: National Health Interview Survey 2013,
2015, and 2017. Epilepsy Behav 2021;122:108194.
[1] Watson NF, Badr MS, Belenky G, Bliwise DL, Buxton OM, Buysse D, et al. Joint
[30] Bergmann M, Tschiderer L, Stefani A, Heidbreder A, Willeit P, Högl B. Sleep
consensus statement of the American academy of sleep medicine and sleep
quality and daytime sleepiness in epilepsy: systematic review and meta-
research society on the recommended amount of sleep for a healthy adult:
analysis of 25 studies including 8,196 individuals. Sleep Med Rev
methodology and discussion. Sleep 2015;38(8):1161–83.
2021;57:101466.
[2] Institute of Medicine Committee on Sleep, M. and Research. The National
[31] Yazdi Z, Sadeghniiat-Haghighi K, Naimian S, Zohal MA, Ghaniri M. Prevalence
Academies Collection: reports funded by National Institutes of Health. In:
of sleep disorders and their effects on sleep quality in epileptic patients. Basic
Colten HR, Altevogt BM, editors. Sleep disorders and sleep deprivation: an
Clin Neurosci 2013;4(1):36–41.
unmet public health problem. Washington (DC): National Academies Press;
[32] Lehner J, Frueh JS, Datta AN. Sleep quality and architecture in Idiopathic
2006.
generalized epilepsy: a systematic review and meta-analysis. Sleep Med Rev
[3] Medic G, Wille M, Hemels ME. Short- and long-term health consequences of
2022;65:101689.
sleep disruption. Nat Sci Sleep 2017;9:151–61.
[33] Fang H, Tu S, Sheng J, Shao A. Depression in sleep disturbance: a review on a
[4] van Golde EG, Gutter T, de Weerd AW. Sleep disturbances in people with
bidirectional relationship, mechanisms and treatment. J Cell Mol Med 2019;23
epilepsy; prevalence, impact and treatment. Sleep Med Rev 2011;15
(4):2324–32.
(6):357–68.
[34] Murphy MJ, Peterson MJ. Sleep disturbances in depression. Sleep Med Clin
[5] Gammino M, Zummo L, Bue AL, Urso L, Terruso V, Marrone O, et al. Excessive
2015;10(1):17–23.
daytime sleepiness and sleep disorders in a population of patients with
[35] Gonzalez-Martinez A, Planchuelo-Gómez Á, Vieira Campos A, Martínez-
epilepsy: a case-control study. J Epilepsy Res 2016;6(2):79–86.
Dubarbie F, Vivancos J, De Toledo-Heras M. Perceived quality of life (QOLIE-
[6] Grigg-Damberger M, Foldvary-Schaefer N. Bidirectional relationships of sleep
31-P), depression (NDDI-E), anxiety (GAD-7), and insomnia in patients with
and epilepsy in adults with epilepsy. Epilepsy Behav 2021;116:107735.
epilepsy attended at a refractory epilepsy unit in real-life clinical practice.
[7] Gibbon FM, Maccormac E, Gringras P. Sleep and epilepsy: unfortunate
Neurol Sci 2022;43(3):1955–64.
bedfellows. Arch Dis Child 2019;104(2):189–92.
[36] Piperidou C, Karlovasitou A, Triantafyllou N, Terzoudi A, Constantinidis T,
[8] Dell KL, Payne DE, Kremen V, Maturana MI, Gerla V, Nejedly P, et al. Seizure
Vadikolias K, et al. Influence of sleep disturbance on quality of life of patients
likelihood varies with day-to-day variations in sleep duration in patients with
with epilepsy. Seizure 2008;17(7):588–94.
refractory focal epilepsy: a longitudinal electroencephalography investigation.
[37] Lee S, Kim JH, Chung JH. The association between sleep quality and quality of
EClinicalMedicine 2021;37:100934.
life: a population-based study. Sleep Med 2021;84:121–6.
[9] Jain SV, Kothare SV. Sleep and epilepsy. Semin Pediatr Neurol 2015;22
[38] Kwan P, Sperling MR. Refractory seizures: try additional antiepileptic drugs
(2):86–92.
(after two have failed) or go directly to early surgery evaluation? Epilepsia
[10] Maltseva M, Schubert-Bast S, Zöllner JP, Bast T, Mayer T, von Spiczak S, et al.
2009;50(Suppl 8):57–62.
Sleep quality, anxiety, symptoms of depression, and caregiver burden among
[39] Strzelczyk A, Griebel C, Lux W, Rosenow F, Reese JP. The burden of severely
those caring for patients with Dravet syndrome: a prospective multicenter
drug-refractory epilepsy: a comparative longitudinal evaluation of mortality,
study in Germany. Orphanet J Rare Dis 2023;18(1):98.
morbidity, resource use, and cost using German health insurance data. Front
[11] Liguori C, Toledo M, Kothare S. Effects of anti-seizure medications on sleep
Neurol 2017;8:712.
architecture and daytime sleepiness in patients with epilepsy: a literature
[40] Lin CY, Wu T, Chang CW, Hsieh HY, Cheng MY, Tseng WJ, et al. Reduced sleep
review. Sleep Med Rev 2021;60:101559.
quality is related to poor quality of life in patients with juvenile myoclonic
[12] Strzelczyk A, Schubert-Bast S. Psychobehavioural and cognitive adverse events
epilepsy, a case-control study. Life (Basel) 2022;12(3).
of anti-seizure medications for the treatment of developmental and epileptic
[41] Chen NC, Tsai MH, Chang CC, Lu CH, Chang WN, Lai SL, et al. Sleep quality and
encephalopathies. CNS Drugs 2022;36(10):1079–111.
daytime sleepiness in patients with epilepsy. Acta Neurol Taiwan 2011;20
[13] Strzelczyk A, Aledo-Serrano A, Coppola A, Didelot A, Bates E, Sainz-Fuertes R,
(4):249–56.
et al. The impact of epilepsy on quality of life: findings from a European
[42] Vendrame M, Yang B, Jackson S, Auerbach SH. Insomnia and epilepsy: a
survey. Epilepsy Behav 2023;142:109179.
questionnaire-based study. J Clin Sleep Med 2013;9(2):141–6.
[14] Gholami A, Salarilak S, Lotfabadi P, Kiani F, Rajabi A, Mansori K, et al. Quality of
[43] Giorelli AS, Passos P, Carnaval T, Gomes MdaM. Excessive daytime sleepiness
life in epileptic patients compared with healthy people. Med J Islam Repub
and epilepsy: a systematic review. Epilepsy Res Treat 2013;2013:629469.
Iran 2016;30:388.

9
C. Lawthom, A. Didelot, A. Coppola et al. Epilepsy & Behavior 148 (2023) 109481

[44] Saad M, Gomaa M, Belal T, Fathi W, Salama S. Sleep architecture in patients [46] Fazekas B, Megaw B, Eade D, Kronfeld N. Insights into the real-life experiences
with idiopathic epilepsy. Sleep Vigilance 2019;3(1):33–8. of people living with epilepsy: a qualitative netnographic study. Epilepsy
[45] Foldvary-Schaefer N. Sleep complaints and epilepsy: the role of seizures, Behav 2021;116:107729.
antiepileptic drugs and sleep disorders. J Clin Neurophysiol 2002;19 [47] Körbel K, Rosenow F, Maltseva M, Müller H, Schulz J, Tsalouchidou PE, et al.
(6):514–21. Impact of COVID-19 pandemic on physical and mental health status and care
of adults with epilepsy in Germany. Neurol Res Pract 2022;4(1):44.

10