Epilepsy & Behavior 78 (2018) 100–103

Contents lists available at ScienceDirect

Epilepsy & Behavior

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

Mozart's music in children with drug-refractory epileptic

encephalopathies: Comparison of two protocols
Giangennaro Coppola a,⁎, Francesca Felicia Operto a, Francesca Caprio a, Giuseppe Ferraioli a, Simone Pisano a,
Andrea Viggiano b, Alberto Verrotti c
a
Clinic of Child and Adolescent Neuropsychiatry, Department of Medicine and Surgery, S. Giovanni di Dio and Ruggi d'Aragona Hospital, University of Salerno, Italy
b
Physiology Institute, Department of Medicine and Surgery, S. Giovanni di Dio and Ruggi d'Aragona Hospital, University of Salerno, Italy
c
Clinic of Pediatrics, University of L'Aquila, Italy

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

Article history: In this prospective, randomized, open label study, we compared the effect on seizure recurrence and quality-of-life
Received 29 June 2017 parameters, of two different protocols of music therapy in children and adolescents with refractory epileptic
Revised 21 September 2017 encephalopathies. Nine out of 19 patients (13 males and 6 females, aged between 1 and 24 years) were randomized
Accepted 28 September 2017
to listen to Mozart's sonata in D major for two pianos K448 for 2 h/day for 2 weeks; other 10 children were
Available online 22 December 2017
randomized on a set of Mozart's compositions.
Keywords:
In group 1 (K448), 2/9 children (22.2%) had a ≥75% seizure decrease; two patients had less than 50% seizure
Mozart's K448 reduction, and the other five were unchanged.
Mozart's set In group 2 (set Mozart), 7/10 patients (70%) had a significant seizure reduction (specifically, ≥50% in 1/10; ≥75%
Music therapy in 4/10; 100% in 2/10).
Childhood epileptic encephalopathy An overall more significant behavioral improvement including less irritability and tearfulness, reduced self-/
Refractory epilepsy heteroaggression, a better daytime vigilance, and nighttime sleep quality, was also reported in children from
group 2.
In conclusion, the present study seems to confirm that music therapy may be an additional, nonpharmacological,
effective treatment for patients with refractory epileptic seizures in childhood. The Mozart's set of different com-
positions can be better accepted and effective than the K448.
© 2017 Elsevier Inc. All rights reserved.

1. Introduction a prospective, randomized, open label study comparing two different

protocols. Following Tomatis' method [10], Mozart's K448 sonata or
In recent years, musical stimulation, in particular the “Mozart effect” Mozart's set of compositions were compared on their effect on seizure re-
of the K448 sonata for two pianos, has been reported to decrease both currence and quality-of-life parameters, including nocturnal sleep and
interictal electroencephalographic (EEG) discharges [1–3] and recurrence mood/behavioral changes, in children and adolescents with refractory
of clinical seizures [4–8]. More recently, a set of Mozart's compositions epileptic encephalopathies [9].
was reported to be particularly beneficial in reducing seizure recurrence
in children and adolescents with drug-resistant structural or genetic epi- 2. Materials and methods
leptic encephalopathies. Improved nocturnal sleep quality together with
positive behavioral and mood changes were also reported. Furthermore, Patients were recruited from the center for children and adolescent
children appeared more compliant to a set of different music proposals epilepsy care of the Medical School of the University of Salerno. The pro-
than a single composition like Mozart's K448 [9]. Nevertheless, in so far tocol was approved by the Ethics Committee, and the study was not
available studies, there is a significant diversity with respect to protocols sponsored by any commercial organization. They were selected accord-
of music therapy as well as for kind of musical compositions, length of ing to the following criteria: (i) 12 or more months of age; (ii) with sei-
daily music listening, and duration of follow-up. Therefore, we conducted zures refractory to antiepileptic drugs; (iii) with drug-resistant epilepsy
and with at least four seizures a week during the 6 months before music
therapy was administered; (iv) without systemic or progressive neuro-
⁎ Corresponding author at: Clinic of Child and Adolescent Neuropsychiatry, Department
of Medicine and Surgery, S. Giovanni di Dio and Ruggi d'Aragona Hospital, University of
logic diseases (including deafness); and (v) with informed consent by
Salerno, Italy. parents and/or caregivers to participate in the study. Exclusion criterion
E-mail address: [email protected] (G. Coppola). was poor compliance with following the study protocol. Nonepileptic

https://doi.org/10.1016/j.yebeh.2017.09.028
1525-5050/© 2017 Elsevier Inc. All rights reserved.
 G. Coppola et al. / Epilepsy & Behavior 78 (2018) 100–103 101

seizures were excluded through video-EEGs and/or long-term monitor- 3. Results

ing EEGs. Brain computed tomography/magnetic resonance imaging
scans were performed in all cases. Seizure frequency, type, and duration According to the inclusion criteria, 19 children (13 males and 6 fe-
were recorded by parents and caregivers, both at home and at school, in males) were recruited, aged between 1 and 24 years at the beginning
an epilepsy diary. Seizures were classified according to the International of the administration of music therapy with a mean age of 14.2 years
League Against Epilepsy (ILAE) classification of epileptic seizures [11]. (group 1) and 12.1 years (group 2). All patients were suffering from
The effectiveness of music therapy was rated as follows: seizure-free drug-resistant epileptic encephalopathies or syndromes.
(100% remission); very good (50–98% decrease in seizure frequency); Diagnoses were as follows: multifocal encephalopathy with spasms
minimal (seizure frequency less than 50% with minimal change in and tonic seizures (n = 16); Lennox–Gastaut syndrome (n = 2); and
seizure severity); and unmodified or worsened (seizure frequency and secondarily generalized cryptogenic partial epilepsy (n = 1). All pa-
severity similar to [unmodified] or worse than [worsened] baseline). tients but one (18/19) had a severe/profound degree of intellectual dis-
Before starting the treatment (TIME 0), parents/caregivers of each ability associated, in 13 out of 18, with a severe type of cerebral palsy
patient were administered a questionnaire, designed to assess the (spastic quadriplegia with or without dystonia, n = 11; double
following variables: seizure frequency and type, quality of nocturnal hemiparesis, n = 1; right hemiplegia, n = 1). One patient had a mild in-
sleep, daytime mood and behavior (irritability, fits of rage, crying spells, tellectual disability.
self-/heteroaggression), and gastrointestinal function (i.e., constipation, The mean number of the antiepileptic drugs was 3.3 (group 1) and
abdominal pain attacks), throughout the 6 months prior to treatment 3.2 (group 2), combined with the ketogenic diet in 4 children.
onset. Laboratory assessment including antiepileptic drug blood level, Seizure recurrence before music therapy was more than 1/day in 16/
full blood count, serum alanine aminotransferase (ALT), aspartate 19 patients, more than 1/week in 3/19.
aminotransferase (AST), gamma-glutamyl transferase (GGT), urea, Nine out of 19 patients were randomized to listen to Mozart's sonata
creatinine, ammonium, and urinalysis, together with a sleep–wake in D major for two pianos K448 for 2 h/day for 2 weeks; the other 10
video-EEG recording, was performed in each patient in the previous children were randomized on a set of Mozart's compositions (including
week. No. 41, K551 symphony; piano concerto No. 22, K482; Violin Concerto
Soon after, each child was delivered with a device (see later), and No. 1, K207; Violin Concerto No. 4 in D major, K218, cheerful open; Sym-
his/her parents/caregivers had an exhaustive training for home treat- phony No. 46 in C major, KV96, cheerful; Flute Concerto in D major
ment. Parents were also provided a website password where they K314, cheerful open) for 2 h a day for 15 days.
could daily connect in order to download the selected music. Compli- The clinical characteristics of patients randomized to receive the
ance to protocol was thus monitored. Moreover, parents were asked K448 or Mozart set are summarized in Table 1.
to report the type, duration, and frequency of the seizures in a diary, In group 1 (K448), 2/9 children (22.2%) had a N75% seizure decrease,
as well as changes in their child's daily activities and/or behavior and two patient had less than 50% seizure reduction, and the other five were
self-deemed compliance to protocol. unchanged. In this group, parents reported a decreased irritability and
Patients randomized to group 1 were asked to listen to Mozart's so- tearfulness in 2 children, and a better gastrointestinal functioning in an-
nata in D major for two pianos, K448, 2 h a day for 15 days (total number other one.
of hours = 30). Children could listen to the music anytime during the In group 2 (set Mozart), 7/10 patients (70%) had a significant seizure
day, even during sleep. reduction (specifically, N50% in 1/10; N75% in 4/10; 100% in 2/10) (Fig. 1).
The same schedule (2 h a day, for 15 days, 30 h in total) was applied With respect to behavioral effects, parents reported decreased irrita-
to patients randomized to group 2, who had to listen a set of Mozart's bility and tearfulness in 8 patients, increased daytime activity in 7, im-
compositions including: symphony No. 41, K551; piano concerto No. proved sleep habits (decreased time to fall asleep reduction in night,
22, K482; violin concerto No. 1, K207; violin concerto No. 4 in D
major, K218, allegro aperto; symphony No. 46 in C major, KV96, allegro; Table 1
and flute concerto in D major K314, allegro aperto. Characteristics of patients of group 1 (K448) and group 2 (set Mozart).
Both groups listened to the music modified by an electronic device Group 1 (9 pts) Group 2 (10 pts)
called “electronic ear” [10]. The latter, through a system of filters,
Sex 5 males; 4 females 8 males, 2 females
amplifiers, and sophisticated mechanisms of electronic gate, modifies Age (years) mean, range 14.2 (1–24) 12.1 (1–23)
the amplitude of sound frequencies preferably delivering higher sound Mental delay
frequencies (N3000 Hz) which, according to Tomatis, “recharge” the Mild 1 –
cerebral cortex. The electronic ear works on the middle ear through Moderate – –
Severe 5 4
the contraction of the muscles of the hammer and the bracket. The Profound 3 6
sound is transmitted through both air conduction and bone conduction. Cerebral palsy
Through the air, the sound signal reaches the eardrum, whose active Spastic tetraplegia 7 8
vibration stimulates the cochlea in the inner ear. Through bone Double hemiplegia – 1
Spastic dyskinetic – 1
conduction, the sound message is transmitted directly on the skull
Hemiplegia 1 –
from a vibrator placed on the top of the cap. The sound, in this way, Epilepsy type
directly reaches the inner ear, bypassing the eardrum. Epileptic encephalopathy 6 10
Fifteen days later, patients were clinically reassessed, antiepileptic LGS 2 –
drug dosages were evaluated, the music device and the filled-in diary Focal epilepsy ± SG 1 –
Seizure type
regarding the first 15 days of treatment were retrieved.
Spasm 2 1
No changes to the antiepileptic treatment or addition of other drugs, Tonic 5 4
except rescue drugs, were allowed throughout the study. Atypical absences 3 1
Statistical analyses were performed by means of SPSS (SPSS Inc., Clonic – 1
Focal ± SG 2 3
USA, 2006). Data are expressed as mean ± standard deviation (SD).
Seizure frequency
Two-way analysis of variance (ANOVA) was carried out to compare ≥1/day 6 5
the number of seizure attacks before and after listening to Mozart's ≥1/week 3 5
K448 or Mozart set of compositions. The t-test was used to compare Number of AEDs (mean, range) 3.3 (2–4) 3.2 (2–4)
the percentage reduction in seizure frequency between the two groups. Legend: LGS, Lennox–Gastaut syndrome; SG, secondary generalization; AED, antiepileptic
A P-value less than 0.05 was set as significant. drug.
102 G. Coppola et al. / Epilepsy & Behavior 78 (2018) 100–103

An overall more significant behavioral improvement, including less

Percentage of seizure decrease in irritability and tearfulness, reduced self-/heteroaggression, a better
group1 and 2 daytime vigilance, and nighttime sleep quality, was also reported in
6
children from group 2 (8/10 patients in group 2 vs. 3/9 in group 1).
5 We have been then able to replicate and extend findings from our
N.of patients

4 previous study [9], where the same Mozart protocol was given for the
k448
3
first time in patients with refractory epileptic encephalopathies in a se-
Mozart verely impaired intellectual and neurological context. In our previous
2
study, a good efficacy in reducing seizure recurrence as well as improv-
1 ing significant quality-of-life related parameters, such as regularity of
0 sleep, vigilance, and environmental participation, was reported.
100% > 75% > 50% < 50% Unchanged Noteworthy, Mozart's music (K448) had been already applied to
Seizure decrease (%) clinical epilepsy in a few studies, in pediatric patients with idiopathic,
benign epilepsy [12], or in more severe forms, including patients with
status epilepticus [6–8,13] or in relation to the potential effects on
Fig. 1. Percentage of seizure decrease in group 1 and group 2. interictal EEG discharges [1–3].
In patients with severe epilepsy, Lin et al. [7] evaluated the effective-
and early morning awakenings) in 6, decreased constipation in 2, and ness of music therapy (Mozart's K448) in 11 children aged between 2.9
drooling in another one. Worth noting is that 3 parents reported that and 14.4 years, with refractory epilepsy of symptomatic or unknown
their children developed some kind of babbling or communicative be- etiology. After 6 months of treatment, a seizure reduction of N 50% was
haviors a few days after starting music therapy. observed in 72.7% of patients, including 2 who were seizure-free. No
The total number of seizures before and after music therapy in both correlation was reported with intelligence quotient (IQ), etiology, and
groups is shown in Fig. 2. The two-way ANOVA with a within-subject sex. The best response was obtained in generalized seizures.
factor time (before vs. after) and a between-subject factor group In the only randomized controlled trial, conducted by Bodner et al.
(group 1 vs. group 2) demonstrated a significant effect for the time [8], the music therapy (Mozart's K448) was administered to an
(F1,32 = 4.15; P b 0.05) but not for the treatment or the interaction. Com- institutionalized adult population with refractory epilepsy and mental
pared with T-0, the reduction in the number of seizures was larger for the retardation followed up for 3 years. Music listening led to a 24%
Mozart-set group (37 ± 13%) than for the K448 group (77 ± 13%); the reduction of the total seizure recurrence compared with baseline period
Student's t-test demonstrated a significant difference between the two and a “carryover” effect after discontinuation of − 33% of the total
groups (P b 0.05). seizure number.
Interestingly, a significant effect on seizure recurrence was reported
4. Discussion in a few adults with focal-onset drop attacks [4,5] and in two patients
with nonconvulsant status epilepticus [6,13].
This randomized, prospective, open label study comparing two pro- Both in the present study and in the previous one [9], the clinical
tocols of music therapy showed that listening to a Mozart's music set of context appears to be much more severe than that of other series for
compositions (group 2), according to the principles of the Tomatis both seizure frequency and resistance and the presence of severe co-
method resulted in a seizure reduction significantly higher than that ob- morbidities such as intellectual disability and/or cerebral palsy.
served with a protocol of listening to Mozart's K448 sonata for two With regard to the kind of protocol used in music therapy, it should
pianos (group 1) (37 ± 13% vs. 77 ± 13%, respectively). A relevant effect be noted their considerable variability in the few works, in relation to
(greater than 50%) was also observed in a greater number of patients the type of musical compositions, the daily duration, and the length of
from the Mozart's set (70%) than from the K448 group (22.2%). Despite the protocol. Mozart K448 was administered by Lin et al. [7] each
the ANOVA failing to demonstrate a significant effect for the interaction night for 8 min for six months, while Bodner et al. [8] administered it
treatment × time, the present result represents a preliminary data that for the full night sleep, at regular intervals of 8.5 min each hour for
encourage the initiation of further studies on a larger scale. In fact, at one year.
least, a significant effect of music therapy has been observed in the The protocol employed (Mozart's music set) in the previous work by
present study with both the protocol used, with a trend toward a possi- Coppola et al. [9] for 2 h a day was suggested by the need of reducing the
ble greater effect for the Mozart's music set protocol. potential monotony due to listening to the same composition for 2 h, al-
though remaining within the harmonics of Mozart's music. This study
80 seems to confirm some greater compliance and effectiveness of this pro-
tocol than the K448 alone, at least in children with a more severe clinical
Total seizure number/2 weeks

70 condition.
60
5. Limitations and strengths of the study
50
K448 The small sample size and the short duration of follow-up may rep-
40
resent the main limitations of the study.
30 Mozart set In addition, data regarding the behavioral effects of music therapy
were gathered by means of a questionnaire for parents and not by
20
adhoc validated scales. Considering the positive effects on behavior
10 and quality of life reported in different series, this issue is worthy to
be further analyzed.
0 The homogeneity of the sample (all children with refractory epilep-
T-0 T-1 tic encephalopathy/epilepsy in comorbidity with intellectual disability/
cerebral palsy), the randomization of children on two protocols of music
Fig. 2. Total seizure number/2 weeks before and after music therapy. Legend: T-0, before therapy, the use for both arms of music strings with high harmonicity
music therapy; T-1, after 2-week treatment. (N 3000 Hz) according to the principles of Tomatis method, based on
 G. Coppola et al. / Epilepsy & Behavior 78 (2018) 100–103 103

the electronic ear, the close monitoring of each patient for study [2] Lin LC, Lee WT, HC Wu, Tsai CL, Wei RC, Jong YJ, et al. Mozart K.448 and epileptiform
discharges: effect of ratio of lower to higher harmonics. Epilepsy Res 2010;89:
compliance by detecting music listening on the website, as well as re- 238–45.
cording the behavioral changes and sleep quality on daily clinical diaries [3] Lin LC, Lee WT, Wu HC, Tsai CL, Wei RC, Mok HK, et al. The long-term effect of
and, finally, an unmodified antiepileptic drug regimen throughout the listening to Mozart K.448 decreases epileptiform discharges in children with
epilepsy. Epilepsy Behav 2011;21:420–4.
different phases of the study may represent potential strengths. [4] Hughes JR, Fino JJ, Melyn MA. Is there a chronic change of the “Mozart effect” on
In conclusion, the present study seems to confirm that music thera- epileptiform activity? A case study. Clin Electroencephalogr 1999;30:44–5.
py may be an additional, nonpharmacological, effective treatment for [5] Lahiri N, Duncan JS. The Mozart effect: encore. Epilepsy Behav 2007;11:152–3.
[6] Miranda M, Kuester G, Ríos L, Basaez E, Hazard S. Refractory nonconvulsive status
patients with refractory epileptic seizures in childhood, in comorbidity epilepticus responsive to music as an add-on therapy: a second case. Epilepsy
with intellectual disability/cerebral palsy. In addition, the Mozart set Behav 2010;19:539–40.
of different compositions can be better accepted and effective than the [7] Lin LC, Lee WT, Wang CH, Chen HL, Wu HC, Tsai CL, et al. Mozart K.448 acts as a
potential add-on therapy in children with refractory epilepsy. Epilepsy Behav
K448 only.
2011;20:490–3.
In addition to seizure recurrence, improvements can be observed [8] Bodner M, Turner RP, Schwacke J, Bowers C, Norment C. Reduction of seizure
with respect to the quality of life, including behavior disorders and occurrence from exposure to auditory stimulation in individuals with neurological
nighttime sleep instability. handicaps: a randomized controlled trial. PLoS One 2012;7:e45303.
[9] Coppola G, Toro A, Operto FF, Ferrarioli G, Pisano S, Viggiano A, et al. Mozart's music
Further studies are, however, necessary for the search of other types in children with drug-refractory epileptic encephalopathies. Epilepsy Behav 2015
of musical compositions and the most appropriate duration of daily and Sep;50:18–22.
long-term music stimulation; the use of different devices as well as the [10] Thompson BM, Andrews SR. An historical commentary on the physiological effects
of music: Tomatis, Mozart and neuropsychology. Integr Physiol Behav Sci 2000;35:
effects of music therapy on the interictal EEG discharges during wake- 174–88.
fulness and sleep deserve to be better understood. [11] Commission on Classification and Terminology of the International League Against
Epilepsy. Proposal for revised clinical and electroencephalographic classification of
epileptic seizures. Epilepsia 1981;22:489–501.
Conflict of interest [12] Lin LC, Lee MW, Wei RC, Mok HK, Yang RC. Mozart K.448 listening decreased seizure
recurrence and epileptiform discharges in children with first unprovoked seizures: a
All authors declare the absence of any potential conflicts of interest. randomized controlled study. BMC Complement Altern Med 2014;13:14–7.
[13] Kuester G, Rios L, Ortiz A, Miranda M. Effect of music on the recovery of a patient
with refractory nonconvulsive status epilepticus. Epilepsy Behav 2010;18:491–3.
References
[1] Turner RP. The acute effect of music on interictal epileptiform discharges. Epilepsy
Behav 2004;5:662–8.