Live music is beneficial to preterm infants in the neonatal intensive care unit environment.

Abstract

Music stimulation has been shown to provide significant benefits to preterm infants. We hypothesized that live music therapy was more beneficial than recorded music and might improve physiological and behavioral parameters of stable preterm infants in the neonatal intensive care unit.
Thirty-one stable infants randomly received live music, recorded music, and no music therapy over 3 consecutive days. A control of the environment noise level was imposed. Each therapy was delivered for 30 minutes. Inclusion criteria were postconceptional age > or = 32 weeks, weight > or = 1,500 g, hearing confirmed by distortion product otoacoustic emissions (DPOAEs), and no active illness or documentation of hyperresponsiveness to the music. Heart rate, respiratory rate, oxygen saturation, and a behavioral assessment were recorded, every 5 minutes, before, during, and after therapy, allowing 30 minutes for each interval. The infant's state was given a numerical score as follows: 1, deep sleep; 2, light sleep; 3, drowsy; 4, quiet awake or alert; 5, actively awake and aroused; 6, highly aroused, upset, or crying; and 7, prolonged respiratory pause > 8 seconds. The volume range of both music therapies was from 55 to 70 dB. Parents and medical personnel completed a brief questionnaire indicating the effect of the three therapies.
Live music therapy had no significant effect on physiological and behavioral parameters during the 30-minute therapy; however, at the 30-minute interval after the therapy ended, it significantly reduced heart rate (150 +/- 3.3 beats/min before therapy vs 127 +/- 6.5 beats/min after therapy) and improved the behavioral score (3.1 +/- 0.8 before therapy vs 1.3 +/- 0.6 after therapy, p < 0.001). Recorded music and no music therapies had no significant effect on any of the tested parameters during all intervals. Both medical personnel and parents preferred live music therapy to recorded music and no music therapies; however, parents considered live music therapy significantly more effective than the other therapies.
Compared with recorded music or no music therapy, live music therapy is associated with a reduced heart rate and a deeper sleep at 30 minutes after therapy in stable preterm infants. Both recorded and no music therapies had no significant effect on the tested physiological and behavioral parameters.

Full text

Page 1

Live Music Is Beneficial to Preterm
Infants in the Neonatal Intensive Care
Unit Environment
Shmuel Arnon, MD, Anat Shapsa, RN, Liat Forman, RN, Rivka Regev, MD, Sofia Bauer, MD,
Ita Litmanovitz, MD, and Tzipora Dolfin, MD
ABSTRACT: Background: Music stimulation has been shown to provide significant benefits
to preterm infants. We hypothesized that live music therapy was more beneficial than
recorded music and might improve physiological and behavioral parameters of stable preterm
infants in the neonatal intensive care unit. Methods: Thirty-one stable infants randomly
received live music, recorded music, and no music therapy over 3 consecutive days. A control
of the environment noise level was imposed. Each therapy was delivered for 30 minutes.
Inclusion criteria were postconceptional age ‡ 32 weeks, weight ‡ 1,500 g, hearing con-
firmed by distortion product otoacoustic emissions (DPOAEs), and no active illness or
documentation of hyperresponsiveness to the music. Heart rate, respiratory rate, oxygen
saturation, and a behavioral assessment were recorded, every 5 minutes, before, during, and
after therapy, allowing 30 minutes for each interval. The infant’s state was given a numerical
score as follows: 1, deep sleep; 2, light sleep; 3, drowsy; 4, quiet awake or alert; 5, actively
awake and aroused; 6, highly aroused, upset, or crying; and 7, prolonged respiratory
pause > 8 seconds. The volume range of both music therapies was from 55 to 70 dB.
Parents and medical personnel completed a brief questionnaire indicating the effect of the
three therapies. Results: Live music therapy had no significant effect on physiological and
behavioral parameters during the 30-minute therapy; however, at the 30-minute interval after
the therapy ended, it significantly reduced heart rate (150 ± 3.3 beats/min before therapy vs
127 ± 6.5 beats/min after therapy) and improved the behavioral score (3.1 ± 0.8 before
therapy vs 1.3 ± 0.6 after therapy, p < 0.001). Recorded music and no music therapies had
no significant effect on any of the tested parameters during all intervals. Both medical
personnel and parents preferred live music therapy to recorded music and no music therapies;
however, parents considered live music therapy significantly more effective than the other
therapies. Conclusions: Compared with recorded music or no music therapy, live music
therapy is associated with a reduced heart rate and a deeper sleep at 30 minutes after therapy
in stable preterm infants. Both recorded and no music therapies had no significant effect on
the tested physiological and behavioral parameters. (BIRTH 33:2 June 2006)
Shmuel Arnon, Anat Shapsa, Liat Forman, Rivka Regev, Ita
Litmanovitz,SofiaBauer,andTziporaDolfinarefromtheDepartment
of Neonatology, Meir Medical Center, Kfar-Saba, Sackler Faculty of
Medicine,TelAvivUniversity,TelAviv,Israel.ElianaGiladandSunita
Staneslow are from the Voices of Eden ensemble, Amirim, Israel.
Eliana Gilad, BA, and Sunita Staneslow, MM, were musicians
who participated in this study.
Address correspondence to Shmuel Arnon, MD, Deputy Director of
the Department of Neonatology, Meir Medical Center, Kfar-Saba,
44281 Israel.
? 2006, Copyright the Authors
Journal compilation ? 2006, Blackwell Publishing, Inc.
BIRTH 33:2 June 2006
131

Page 2

The prognosis of infants with very low birthweight
has significantly improved during the last decade,
without concomitant improvement in neurodevelop-
mental outcome (1). The importance given to stress
reduction techniques, to allow optimal neurobeha-
vioral function, led to the development of a range
of therapies such as the Neonatal Individualized
DevelopmentalCareand
(NIDCAP) (2); control of external stimuli (vestibular,
auditory, visual, tactile); clustering of nursery care
activities; and positioning or swaddling of preterm
infants to provide a sense of containment similar to
the intrauterine experience (3).
Music was also suggested as a tool for alleviating
stress, providing a cost-effective and enjoyable strat-
egy to improve empathy and compassion without
interfering with routine work in the hospital and neo-
natal intensive care unit setting (4,5). A meta-analysis
of 10 clinical trials showed that music therapy has a
significant clinical benefit for preterm infants in the
neonatal intensive care unit (6). A positive effect on
long-term variables such as length of hospitalization,
weight gain, and nonnutritive sucking rate, and short-
term variables such as oxygenation, heart rate, and
behavioral score was reported (7).
Most studies used recorded music delivered by free
field or earphones, and compared the yield of lulla-
bies with routine auditory stimulation (8–10). (‘‘Free
field’’ delivery means that the sound is transferred to
the ear via sound waves and not via artificial con-
ductor, that is, earphone.) However, the effect of
pop, jazz, and other Western types of music versus
lullaby and classical music, as well as the effect of live
versus recorded music delivery systems, were not sys-
tematically assessed. The purpose of our study was to
compare the effect of live music therapy, recorded
music therapy, and no music therapy on short-term
physiological and behavioral parameters of stable
preterm infants during their stay in the neonatal
intensive care unit.
Assessment Program
Methods
Participants
The study was carried out in the neonatal intensive
care unit of Meir Medical Center, Kfar-Saba, Israel,
during a 6-month period that ended in March 2005.
Preterm infants were assigned to receive live music
therapy, the same recorded music therapy, or no
music therapy. All therapies were performed in a
random order over 3 consecutive days for 30 minutes,
to 3 infants at a time, starting an hour after comple-
tion of feeding.
Inclusion criteria were postconceptional age ‡ 32
weeks, weight ‡ 1,500 g, hearing confirmed by dis-
tortion product otoacoustic emissions (DPOAEs),
and no acute illness or documentation of hyperre-
sponsiveness to the music. Infants were excluded if
they had anomalies associated with neurological pro-
blems (i.e., intraventricular hemorrhage grade 3–4,
periventricular leukomalacia) or received medications
such as phenobarbital or oxygen that might interfere
with the reaction to music stimuli. The study protocol
was approved by the Institutional Review Board, and
a written informed consent was obtained from all
parents.
Measures
Heart rate (beats per minute), respiratory rate, oxy-
gen saturation, and the infant’s behavioral state as
defined by Als (11) were recorded, every 5 minutes, in
3 intervals; before, during, and after the 3 therapies,
allowing 30 minutes for each interval. The infant’s
state was given a numerical score as follows: 1, deep
sleep; 2, light sleep; 3, drowsy; 4, quiet awake or alert;
5, actively awake and aroused; 6, highly aroused,
upset, or crying; and 7, prolonged respiratory
pause > 8 seconds.
In all study phases the objective measurements of
music or background sound levels were performed
with the 407790 Octave Band Sound Analyzer, a
type 2 integrating sound level meter using the
decibel-A scale filter (Extech Instruments, Melrose,
Massachusetts, USA). Before each recording, the
sound level meter that simulated typical reception of
sound in the human ear was adjusted with the 407744
sound level calibrator (Extech Instruments, Melrose,
Massachusetts, USA), according to the manufac-
turer’s instructions and specifications at a range of
30 to 130 dB and a frequency of 25 to 10,000 Hz.
After each therapy interval, both medical person-
nel and parents were asked to complete a brief ques-
tionnaire indicating the effect of the therapy. Each
interval was graded on a scale of 1 to 10; a grade of 4
or less indicated a harmful effect, 5 indicated no
effect, and 6 and more indicated a beneficial effect.
Procedure
The infant was lying in a supine position, and music
was played at a volume between 55 and 70 dB. Live
music was performed at a distance of 1 to 2 meters
from the infant’s bed unit and comprised a lullaby
style, soothing, rhythmic, repetitive, wordless blend
of Eastern and Western musical elements sung by
female voice with frame drum (E.G.) and an
132
BIRTH 33:2 June 2006

Page 3

accompanying instrument (harp, S.S.). The same
music was played by means of a free field system,
while a tape recorder with two speakers was placed 1
meter from the infant’s bed unit.
For all 3 therapies, a control of the environment
noise level was imposed to maintain constancy and to
minimize possible effects on emotions and behavior
of the infants. The monitor alarms were silent and the
ward’s door was closed to minimize interference from
the outside. As recommended by the American
Academy of Pediatrics (12), background noise mea-
sured near the infant’s ear did not exceed 45 dB. All
therapies were carried out an hour after the comple-
tion of feeding, while infants were on a 3-hour feed-
ing schedule. Throughout the study period, the same
nurses (A.S., and L.F), who received special training
to define the infants’ behavioral state, collected the
data.
Data Analysis
A
measures analysis was designed. The medical staff
performed data analysis in a blinded fashion, each
individual being unaware which infant received each
therapy. Statistical analysis was performed using the
GBStat v 6.5 PC software (Dynamic Microsystems
Inc, Silver Spring, Maryland, USA). The multiple t
test comparisons with Bonferroni adjustment and the
Kruskal-Wallis ANOVA test were used to compare
the medians of the 3 therapy groups. Unless other-
wise indicated, data were expressed as mean ± stan-
dard deviation. Probability values of less than 0.05
were considered significant.
within-subjects,counterbalancing,repeated-
Results
Of the 37 preterm infants who were eligible for enter-
ing the study, 31 were enrolled. Three were excluded
because of incomplete data and 3 others because
parents refused participation in all 3 therapies. No
infants had signs of hyperresponsiveness to the
music. Characteristics of the study sample are
shown in Table 1.
The mean sound level of the 2 music therapies was
not significantly different; both played at a higher
volume than the background noise and not exceeding
70 dB (Table 2).
The heart rate, oxygen saturation, respiratory
rate, and behavioral assessment score of the
infants in all 3 therapies were not statistically
different 30 minutes before therapy. During the
30-minute therapy, live music therapy was asso-
ciated with a decrease of the heart rate that was
the lowest after 20 minutes; however, this decrease
was statistically insignificant. At the 30-minute
interval after therapy, live music therapy was
associated with a significantly reduced heart rate,
with the lowest rate after 24 minutes (p < 0.01)
(Table 3; Fig. 1). The behavioral assessment score
Table 1. Study Sample Characteristics (n = 31)
CharacteristicMedian (Range)NumberFrequency
Male/female
Jews/Arabs
Gestational age (wk)
Birthweight (g)
Apgar score (1 min)
Apgar score (5 min)
Postconceptional age (wk)
Weight at testing (g)
Chronic lung disease
Intraventricular hemorrhage grade 1–2
14/17
18/13
29 (25–34)
1175 (650–1737)
4 (0–9)
7 (4–10)
34 (32–40)
1738 (1584–2524)
4/31
6/31
170
160
150
140
Heart Rate (BPM)
Before
0 102030 40 50
Time (Minutes)
6070 80 90
LMT
RMT
NMT
*
100
DuringAfter
130
120
110
Fig. 1. Effect of each therapy on the infant’s heart rate
(p < 0.001, between the after-live music therapy
[LMT] interval and all other intervals pertaining to
live music interval, recorded music therapy, and no
music therapy).
BIRTH 33:2 June 2006
133

Page 4

showed a gradual decline while the infant was
receiving live music therapy, but it reached statis-
tical significance only after therapy with the low-
est score at 25 minutes (p < 0.001) (Fig. 2). No
infant experienced prolonged respiratory pause
greater than 8 seconds. Lower respiratory rate
and increase oxygen saturation were also docu-
mented in the after-therapy interval compared
with the two other intervals; however these differ-
ences did not reach statistical significance (Table
3). No physiological or behavioral effects were
recorded inrecordedmusic
music therapy during all 3 intervals.
Both medical personnel (n = 23) and parents
(n = 26) preferred live music therapy to recorded
music and no music therapies (p = 0.07 and p = 0.04,
respectively); however, only among the parents was live
musictherapysignificantlymoreeffectivethanrecorded
and no music therapies. None of the 3 therapies was
considered ineffective or harmful (Table 4).
therapy andno
Discussion and Conclusions
Preterm infants are vulnerable to a range of diseases
associated with immaturity of the organ systems and
to stress stimuli during their stay in the neonatal
intensive care unit. A favorable environment may
help to reduce some of these morbidities. The effect
of music was shown to have a positive effect on care-
giver attitudes, mood, and behavior and, concomi-
tantly, on the infants’ physiological and behavioral
state (13). Our study design elements mimic sounds
that the infant may have been exposed to in the uterus
and the mother’s heart beat by using a wordless,
rhythmic, constant, repetitive, soothing lullaby sung
by a female voice with Eastern frame drum and an
accompanying instrument. Being a multicultural
society, our infants mainly belong to Jewish and
Arab families with different cultural and musical back-
grounds; therefore, a blend of Eastern and Western
wordless melody that can apply to all was chosen.
The musical duet adjusted well to the neonatal
intensive care unit space without interfering with tech-
nical aspects of care or routine work, and provided a
valuable tool for alleviating stress. Our data showed
that live music played in the neonatal intensive care
unit to stable preterm infants at 32 weeks postconcep-
tional age and older for 30 minutes resulted in an
improvement of physiological and behavioral short-
term stress parameters. Live music therapy was asso-
ciated with a significant decrease in heart rate and a
calmer deeper sleep, occurring only 30 minutes after
therapy ended. Respiratory rate and oxygen satura-
tion also improved after live music therapy, but these
parameters did not reach statistical significance.
In contrast to live music therapy, both recorded
music therapy and no music therapy had no bearing
on the tested parameters. Standley found that the
combined effect of live singing (Brahms’ lullaby)
and multimodal stimulation was associated with
increased weight gain per day and an earlier dis-
charge of female infants compared with the group
who received no stimulation (7). In another study,
infants whose parents received training in music and
multimodal stimulation had improved parent-infant
bonding, shorter length of hospitalization, and daily
greater average of weight gain (13). Neither study
compared live with recorded music or evaluated the
short-term effect of music. Considering that weight
gain and time to discharge may be influenced by
factors other than music therapy, the net effect of
music can be judged more precisely while analyzing
physiological and behavioral short-term parameters.
Furthermore, in our within-subjects, repeated-
measures design conducted for 3 consecutive days,
Table 2. Comparison Between the Mean Music and No
Music Sound Levels During the 3 Study Therapies
Therapy
Mean Sound
Level (dB)*
Range of Mean
Sound Level (dB)
LMT
RMT
NMT
60.35 ± 4.85
63.72 ± 5.12
40.61 ± 3.63
55.53–64.79
54.66–69.21
38.27–44.72
*Mean ± SD.
LMT = live music therapy; RMT = recorded music therapy;
NMT = no music therapy.
Table 3. Effect of Live Music Therapy on Physiological and Behavioral Parameters in Preterm Infants During Study Therapies
(n = 31)*
Test ParameterBefore Therapy During TherapyAfter Therapy
Heart rate (BPM)
Respiratory rate (per min)
O2saturation
Behavioral score
150 ± 3.3
49 ± 7.3
91 ± 6.8
3.1 ± 0.8
144 ± 3.7
52 ± 8.3
89 ± 10.1
2.5 ± 0.3
127 ± 6.5†
41 ± 7.9
94 ± 4.5
1.3 ± 0.6†
*Mean ± SD; † p < 0.05, between after and before therapy and after and during therapy. BPM = beats per minute.
134
BIRTH 33:2 June 2006

Page 5

variations among participants, such as difference of
postconceptional age, history of illness, or change in
environmental variables within the study period,
which may be presented in a longer period, were
eliminated.
Ourresultsshowedthatlivemusicwasconsideredby
parents to be significantly more beneficial than
recorded music and no music therapies. As for medical
personnel, all 3 therapies were considered beneficial for
the preterm infant, with a tendency for live music
therapy. Other investigators showed that medical per-
sonnel in the neonatal intensive care unit preferred
recorded music to live music by more than twofold
(14); however, no specification was given as to the
type of music that was played and for how long, or
any other explanation for this preference. We hypothe-
sized that the crooning quality of the human voice
compared with recorded music was reassuring and
calming to both babies and adults, bringing about a
change in the neonatal intensive care unit environment.
Nurses were more relaxed and positive, and their inter-
actions with infants and parents were improved. Live
music may also have some intrinsic sound properties
that are absent in recorded music (5).
Habituation that is defined as a decreasing
response to a repeated stimulus is a fundamental
type of learning that may influence behavioral
responses (15,16). Our type of music was introduced
to the infants for the first time in their life and in a
random order. In addition, since the same music was
played only twice, no habituation effect could con-
found the study results.
It was shown that exposure to live harp music was
beneficial to preterm infants and lowered cortisol
levels and respiratory rates (17). Nevertheless, mea-
suring cortisol levels in preterm infants required
manipulation or even venipuncture and may not be
an accurate sign of stress reduction in the neonatal
intensive care unit environment. Even months after
hospitalization, preterm infants with extremely low
gestational age may exhibit high basal and sustained
levels of cortisol because of chronic stress (18).
Given that the data were recorded for 30 minutes
before, during, and after therapy, no conclusion can
be drawn as to the actual length of the effect of
music. Extending the length of follow-up to an hour
or more to see if the effect was sustained could vali-
date the findings, yet a longer follow-up would inter-
fere with the feeding schedule of the infants.
In conclusion, it was shown that, compared with
recorded music or no music therapies, live music
therapy is associated with a reduced heart rate and
a deeper sleep in stable preterm infants at 30 minutes
after therapy. Both recorded and no music therapies
had no significant effect on the tested physiological
and behavioral parameters. Further research is
needed to assess the feasibility and reproducibility
of our results and their long-term effect on the devel-
opment of preterm infants.
References
1.Bregman J. Developmental outcome in very low birthweight
infants. Current status and future trends. Pediatr Clin North
Am 1998;45:637–690.
Als H, Lawhon G, Brown E, et al. Individualized behavioral
and environmental care for the very low birth weight preterm
infant at high risk for bronchopulmonary dysplasia: neonatal
intensive care unit and developmental outcome. Pediatrics
1986;78:1123–1132.
Symington A, Pinelli J. Developmental care for promoting
development and preventing morbidity in preterm infants.
Cochrane Database Syst Rev 2003;(4):CD001814.
2.
3.
3.0
Awake
Deep Sleep
Before
01020 304050
Time (Minutes)
607080
LMT
RMT
NMT
90100
DuringAfter
2.8
2.6
2.4
2.2
Behavioral Score
2.0
1.8
1.6
1.4
1.2
Fig. 2. Effect of each therapy on the infant’s beha-
vioral state (p < 0.001, between the after-live music
interval [LMT] interval and all other intervals pertain-
ing to live music therapy, recorded music therapy, and
no music therapy).
Table 4. Attitudes of Medical Personnel and Parents
Toward Live Music, Recorded Music, and No Music
Therapies
Therapy
Medical Personnel
(n = 23)
Parents
(n = 26)
LMT
RMT
NMT
8.3 ± 2.1
7.8 ± 1.8
8.0 ± 2.3
9.2 ± 1.4†
8.5 ± 2.2
7.9 ± 2.6
*Mean ± SD; † p < 0.05, between live music therapy and the two
other (recorded music and no music therapies).
LMT = live music therapy; RMT = recorded music therapy;
NMT = no music therapy.
BIRTH 33:2 June 2006
135