Non-nutritive sucking habits in sleeping infants.
ABSTRACT Pacifier use has been postulated to decrease the risk of sudden infant death syndrome (SIDS). The responsible mechanisms are, however, unclear.
Since little is known about the non-nutritive sucking (NNS) habits of infants during sleep, we investigated NNS patterns and changes of physiological parameters during NNS in sleeping infants.
Polygraphic recordings were performed in 12 infants with a median age of 55 days (range 7-82) who regularly used a pacifier during sleep. Episodes of active suckling (bursts) and quiescent periods were differentiated by video observations. We evaluated the time of suckling in relation to the total time of pacifier use, the median number of bursts per min, the median duration of single bursts and the median interval between 2 sequent bursts. In 48 randomly selected bursts, we additionally analyzed changes in heart rate, respiratory frequency and oxygen saturation compared to the 10-second period preceding the burst.
Median sleep time with a pacifier held in mouth was 31.3 min (13.0-117.6), of which 15.5% (6.4-36.7%) was spent with active suckling. The median number of bursts per min was 2.2 (1.2-4.5). The median duration of a burst was 3 s (1-22) and the median interval between 2 bursts was 10 s (1-1,434). Heart rate, respiratory frequency and oxygen saturation did not change significantly during suckling bursts.
This pilot study presents important data for sucking habits in pacifier users which may provide a basis for further investigations concerning the efficacy of pacifiers in SIDS prophylaxis.
- SourceAvailable from: Stephanie YiallourouCurrent Pediatric Reviews 02/2010; 6(1).
- Pädiatrie & Pädologie 02/2010; 45(1):12-15.
Fax +41 61 306 12 34
Non-Nutritive Sucking Habits in Sleeping
Marie Hanzer a Heinz Zotter
Wilhelm Müller a Reinhold Kerbl c
a Werner Sauseng
b Gerhard Pichler
Divisions of a Neonatology and b General Pediatrics, Department of Pediatrics and Adolescent Medicine,
Medical University of Graz, Graz , and c Department of Pediatrics and Adolescent Medicine, Hospital of Leoben,
Leoben , Austria
was 2.2 (1.2–4.5). The median duration of a burst was 3 s (1–
22) and the median interval between 2 bursts was 10 s (1–
1,434). Heart rate, respiratory frequency and oxygen satura-
tion did not change significantly during suckling bursts.
Conclusions: This pilot study presents important data for
sucking habits in pacifier users which may provide a basis for
further investigations concerning the efficacy of pacifiers in
Copyright © 2009 S. Karger AG, Basel
Two different types of sucking are performed by hu-
man infants: nutritive sucking (NS) and non-nutritive
sucking (NNS), also denoted as ‘suckling’. NS occurs ex-
clusively in the presence of oral fluid while NNS takes
place in the absence of fluid. NNS appears in utero, as
confirmed by ultrasound technique  , and represents
one of the first coordinated muscular activities of the fe-
tus [2, 3] . Oral and gag reflexes appear at 12–16 weeks of
gestation, sucking at 24 weeks, and coordinated sucking
and swallowing at 28 weeks. However, a fully coordinated
sucking, swallowing and breathing pattern, which is re-
quired for effective feeding, is not present until 32–34
weeks  .
Non-nutritive sucking ? Pacifier ? Polysomnography ?
Suckling ? Sudden infant death syndrome
Background: Pacifier use has been postulated to decrease
the risk of sudden infant death syndrome (SIDS). The respon-
sible mechanisms are, however, unclear. Objectives: Since
little is known about the non-nutritive sucking (NNS) habits
of infants during sleep, we investigated NNS patterns and
changes of physiological parameters during NNS in sleeping
infants. Methods: Polygraphic recordings were performed
in 12 infants with a median age of 55 days (range 7–82) who
regularly used a pacifier during sleep. Episodes of active
suckling (bursts) and quiescent periods were differentiated
by video observations. We evaluated the time of suckling in
relation to the total time of pacifier use, the median number
of bursts per min, the median duration of single bursts and
the median interval between 2 sequent bursts. In 48 ran-
domly selected bursts, we additionally analyzed changes in
heart rate, respiratory frequency and oxygen saturation
compared to the 10-second period preceding the burst. Re-
sults: Median sleep time with a pacifier held in mouth was
31.3 min (13.0–117.6), of which 15.5% (6.4–36.7%) was spent
with active suckling. The median number of bursts per min
Received: June 23, 2008
Accepted after revision: January 12, 2009
Published online: July 31, 2009
formerly Biology of the Neonate
Heinz Zotter, MD
Division of Neonatology, Department of Pediatrics and Adolescent Medicine
Medical University of Graz , Auenbruggerplatz 30, AT–8036 Graz (Austria)
Tel. +43 316 385 80531, Fax +43 316 385 3300
© 2009 S. Karger AG, Basel
Accessible online at:
Hanzer /Zotter /Sauseng /Pichler /Müller /
NS and NNS differ in terms of organization, duration
and the rate and strength of the sucks [5, 6] . NS is an al-
most continuous behavior, while NNS alternates between
periods of active sucking (bursts) and brief pauses [5, 7] .
The rate of sucking is higher and the pressure amplitude
of sucking is smaller during NNS than during NS  .
Wolff  , who investigated NS and NNS patterns 40 years
ago, revealed a mean frequency of 60 sucks/min during
NS and 120 sucks/min during NNS, suggesting that 2
functionally different systems might be responsible for
If NNS is not stimulated, it disappears by the age of
4–5 months. However, in Western cultures, NNS is regu-
larly released by caregivers and parents who provide their
baby with a pacifier. A number of positive as well as neg-
ative effects of pacifier use have been described in the
literature  . It has been postulated that pacifier use
causes dental malocclusion [2, 9] and an increased risk of
otitis media  . The question whether pacifier use inter-
feres with breastfeeding is discussed controversially. Sev-
eral studies have found that the use of pacifiers is associ-
ated with a shortening of the breastfeeding period, but
the cause and effect relationships are not clear [11, 12] .
Howard et al.  showed that infants who used a paci-
fier very shortly after birth (within the 2nd until 5th day
of life) were breastfed less often at the age of 1 month
when compared to nonusers. However, the authors did
not find such a correlation if the pacifier was introduced
at the age of 1 month. Other studies found no negative
effects of pacifier use on the duration of breastfeeding in
term  and preterm infants  .
Despite these potential adverse effects, many investi-
gators have considered pacifier use as a beneficial inter-
vention. Beside their well-known calming effects, paci-
fiers have been shown to reduce pain in neonates during
painful procedures  . Furthermore, NNS encourages
the development of sucking behavior and improves the
digestion of enteral feeding. Numerous studies have
shown benefits of pacifier use in the transition from ga-
vage to oral feeding in preterm infants, resulting in a sig-
nificant decrease in the length of hospital stay [17–21] .
In recent years, the ‘pacifier debate’ has become even
more important because pacifier use has also been pos-
tulated to decrease the risk of sudden infant death syn-
drome (SIDS) [22–32] .
However, despite many speculations, the mechanisms
by which pacifiers might prevent SIDS remain unclear
[33, 34] .
Since little is known about NNS habits during sleep,
the state in which most SIDS cases occur, it was the aim
of our study to further investigate NNS in healthy sleep-
ing infants using a pacifier. We evaluated the frequency
and duration of suckling bursts, the intervals between
bursts and changes of physiological parameters during
The study was designed as a pilot study. It was performed in
12 infants (6 male, 6 female) who regularly used a pacifier during
sleep. The median age of the infants at the time of the investiga-
tion was 55 days (range 7–82). All subjects were born at term and
their weight was appropriate for their gestational age. All subjects
were free of disorders as assessed by their medical history and
physical examination. No infant was on any medication known
to affect sleep and/or respiratory and cardiac control.
The infants underwent polysomnography in a pediatric sleep
laboratory. Polygraphic recordings were performed during undis-
turbed daytime naps with all infants sleeping in the supine posi-
The study was approved by the local ethics committee. In-
formed consent was obtained from all parents.
A computerized system (Multichannel System, B.E.S.T. Medi-
cal Systems, Vienna, Austria) was employed for all polygraphic
recordings. Polysomnography included the following parameters:
electroencephalogram, which was adapted from the 10–20 system,
using electrodes with gold-gold cups which were placed in order
to permit derivation from C4-O2 and C3-O1. An electrooculo-
gram and electrocardiogram were sampled at a rate of 64 Hz.
Heart rate (HR) was calculated from beat-to-beat ECG measure-
ment. Oxygen saturation was monitored by pulse oximetry (O 2 –
SMO; Novametrix Medical Systems Inc., Wallingford, Conn.,
USA) with an averaging period of 2 s. Capnography was performed
by placing an infant nasal CO 2 probe (Salter Labs, Arvin, Calif.,
USA) with the tips in the infant’s nostrils. Two piezo respiratory-
effort sensors were used to record the infant’s breathing efforts,
placing 1 belt on the chest and 1 on the abdomen to facilitate sep-
arate tracings. For electromyogram recording, surface electrodes
with gold-gold cups were placed on the submental muscle and the
reference electrode on the mandible. Body movements were de-
tected using both video recording and a piezo-crystal actimeter
which was fixed on the right lower leg.
Video recordings were used to differentiate periods in which
the pacifier was held in the infant’s mouth from periods without
All infants were bedded and fell asleep with a pacifier in the
mouth. However, in most cases the pacifier became dislodged
within the first 10 min of the investigation. If the infant continued
to sleep after pacifier dislodgement, it was not put back into the
infant’s mouth. If the infant woke up after the loss of the pacifier
and did not fall asleep again within 5 min, the pacifier was re-
Sucking Habits in Infants
Most infants lost their pacifier several times during the inves-
tigation and received it back due to awakening, causing several
periods with and without a pacifier to be present during poly-
Periods without pacifier and periods of awakening were ex-
cluded from the calculations.
The mouth movements of the infants during active suckling
could be clearly identified by video observation, which facilitated
the distinction between suckling bursts and rest periods in be-
Further analysis included the following parameters:
– active suckling in percent (time of active suckling in relation
to the total time of pacifier use)
– burst frequency (number of bursts per min)
– burst duration
– interval between 2 sequent bursts.
To evaluate potential differences in the frequency of bursts
dependent on sleep state, all suckling bursts were assigned to ei-
ther active sleep, quiet sleep or indeterminate sleep.
Furthermore, 4 suckling bursts following a rest period of at
least 10 s (to provide baseline values) were randomly selected in
each infant. In this total number of 48 bursts, we additionally
analyzed changes in HR, respiratory frequency (RF) and oxygen
saturation compared to the 10-second period preceding the burst
(pre10) using the following formula: Percentage of modifica -
tion = [(mean value during burst – mean value during pre10)/
mean value during pre10] ! 100.
This proportional change was positive if the parameter of in-
terest increased and negative if it decreased.
Analysis of variance was used to compare parameters of inter-
est. Data were shown as means 8 SD except where otherwise in-
p ! 0.05 was considered significant. Calculations were made
using WinStat 3.1 (Kalmia Co. Inc., Cambridge, UK).
The median duration of the polygraphic recording
was 107.6 min (range 44.2–135.7).
After exclusion of awakenings and periods without the
pacifier, 31.3 min (13.0–117.6) remained for analysis of
When subdividing pacifier usage time according to
sleep state, 8.1 min (0.0–60.6) were spent in active sleep,
24.7 min (0.0–55.6) were spent in quiet sleep and 0.8 min
(0.0–7.8) were spent in indeterminate sleep.
Altogether 1,110 suckling bursts were scored.
Of these 1,110 bursts, 414 occurred in active sleep, 630
in quiet sleep and 66 in indeterminate sleep.
The number of bursts per min did not differ signifi-
cantly when separating it according to sleep state. See ta-
ble 1 for details of this and other NNS parameters.
HR, RF and oxygen saturation did not change signifi-
cantly during bursts when compared to the 10-second
period preceding the bursts ( table 2 ).
We investigated NNS habits in healthy sleeping in-
fants with regard to frequency and duration of suckling
bursts, intervals between bursts and changes of physio-
The study was designed as a pilot project. It was per-
formed under the perspective that pacifier use during
sleep has been reported to decrease the risk of SIDS [22–
32] . Thus far, however, there is no proven explanation
how pacifier use might do this.
SIDS is defined as the sudden death of an infant that
is unexpected by history and unexplained by a thorough
postmortem examination, including a case investigation,
complete autopsy and examination of the death scene
 . Despite the dramatic decline of the incidence of
SIDS due to nonprone sleeping campaigns all over the
world, SIDS still remains the leading cause of death in the
Table 1. NNS parameters of 12 full-term infants during sleep
Active suckling, %
Burst frequency (bursts per min)
Burst frequency during AS
Burst frequency during QS
Burst frequency during IS
Burst duration, s
Burst interval, s
AS = Active sleep, QS = quiet sleep, IS = indeterminate sleep.
Table 2. HR, RF and oxygen saturation during suckling bursts
and during the 10-second period preceding the burst (pre10)
Data given as means 8 SD.
Hanzer /Zotter /Sauseng /Pichler /Müller /
postneonatal period  . Identification of new preven-
tive measures to further reduce the incidence of SIDS
continues to be a priority. In this connection, numerous
studies discussing the association between pacifier use
and SIDS have been published in recent years.
In 1979, Cozzi et al.  postulated that the use of a
pacifier might protect against SIDS. Support for this hy-
pothesis was first reported by Mitchell et al.  in 1993.
Since then, several research teams have independently re-
ported a significantly reduced risk of SIDS in pacifier us-
ers [24–32] . Hauck’s meta-analysis, including 7 case-con-
trol studies, revealed a 50% reduced risk of SIDS among
pacifier users during last sleep when compared to a con-
trol group. She found that 1 SIDS death could be prevent-
ed for every 2,733 infants who use a pacifier when placed
for sleep  . Recently, a Californian study confirmed
these findings and even showed a risk reduction of 90%
for pacifier users  . In October 2005, the American
Academy of Pediatrics recommended the use of a paci-
fier throughout the 1st year of life  . In contrast, Mitch-
ell et al.  concluded in their review not to discourage
the use of pacifiers but (due to concerns about possible
adverse effects, especially on breastfeeding) not to spe-
cially recommend it either.
The mechanisms by which pacifiers might prevent
from SIDS remain unclear [33, 34] , especially if consider-
ing that the time an infant keeps the pacifier in his/her
mouth is very short  . It has been hypothesized that
pacifiers might prevent the tongue from sealing off the
airways  , favor mouth breathing in case of nasal oc-
clusion  , discourage the infant from turning to the
prone position  , increase arousability  and reduce
episodes of gastroesophageal reflux  .
Despite the growing interest in the ‘pacifier effect’ re-
garding SIDS in recent years, little is known about the
physiological suckling behavior of infants during sleep,
the state in which most SIDS cases occur. Pollard et al.
 studied night-time NNS habits in infants aged 1–5
months in a sleep laboratory; however, they defined NNS
on a pacifier as having a pacifier in the mouth without
distinguishing between active suckling and rest periods.
Other investigators either studied infants who were awake
or did not differentiate between sleep and wakefulness.
In contrast, suckling parameters were exclusively ana-
lyzed during sleep in our study and awakenings were ex-
cluded from the calculations.
All infants in our study showed the typical rhythmic
NNS pattern, consisting of rapid bursts of suckling activ-
ity and rest periods in between. This behavior has already
been described by Wolff in 1968  . He reported that the
rhythm of suckling is nearly the same in all healthy term
infants and may therefore be considered as a species-spe-
cific mechanism for the temporal regulation of motor be-
havior  . It has been reported that not only term  ,
but also healthy preterm infants were able to suck non-
nutritively in the typical way, even those under 30 weeks
of gestation  . However, with increasing maturation, a
more stable rhythm with a faster pace, higher amplitude
and longer bursts is established, suggesting that a prepro-
grammed rhythm generator is gradually modulated dur-
ing maturation  . Furthermore, gender, state of activ-
ity and weight might influence the suckling pattern in
preterm infants to some extent  .
Since suckling is a complex motor activity, which is
under the control of the brainstem and requires the func-
tion of the cranial nerves V, VII and XII  , neurological
disorders may affect the NNS pattern as well. In Wolff’s
study  , the NNS patterns of infants with various neu-
rological diseases were compared with those of healthy
controls, revealing diverse results dependent on the na-
ture and severity of the underlying disease. However, no
correlation between a specific disorder and a specific ab-
normality in the NNS pattern was found. In contrast,
Hafström and Kjellmer  reported a slower suckling
frequency and a greater variability of suckling frequency
and amplitude in preterm infants who subsequently de-
veloped neurological sequelae.
In our study, the median proportion of suckling in re-
lation to the total time of pacifier use was 15.5%. Haf-
ström et al.  reported a mean suckling activity of
22.7% in full-term infants, whereas healthy preterm in-
fants showed a lower percentage  . However, the term
infants were already investigated on their 2nd or 3rd day
of life, and data were not presented separately for sleep
and wakefulness. Nevertheless, ours as well as Hafström’s
study show that most of the time an infant holds a paci-
fier in his/her mouth is spent without suckling.
Furthermore, we analyzed the duration and frequency
of suckling bursts in our study. The median duration of
bursts was 3 s. Hafström et al.  reported a mean dura-
tion of 3.3 s in term infants and Hafstöm and Kjellmer
 reported shorter bursts in preterm infants. Thus, the
duration of bursts seems to increase with increasing age
or maturity. Median burst frequency, expressed as the
number of bursts per min, was 2.2 in our study. Com-
parisons with other studies are hampered by the fact that
other investigators primarily analyzed the frequency
within the burst, i.e. the number of sucks per burst [7, 43,
44, 46] . Hafström et al.  used a specially designed
computer-based system which facilitated automatic mea-
Sucking Habits in Infants
surements of various NNS parameters. In contrast, all
recordings were analyzed manually in our study.
The median interval between 2 sequent bursts was 10
s in our study. Hafström et al.  reported a mean inter-
val of 5.99 s and Wolff  of 6.61 s; however, in Hafström’s
study all rest periods longer than 20 s were excluded. We
made the observation that most of the infants sucked in
a very regular pattern, consisting of suckling bursts and
brief pauses of just some seconds, during parts of the in-
vestigation; however, there were also periods in which
very long rest periods (the longest lasting for 23.9 min)
occurred, creating a wide range of data.
Since impairment or immaturity of autonomic control
has been postulated as playing a major role in the patho-
genesis of SIDS [47–50] , we also evaluated whether suck-
ling may alter various physiological parameters. Neither
HR nor oxygen saturation changed significantly during
bursts when compared to the 10-second period preceding
the burst. To our knowledge, this aspect has not been in-
vestigated before. Though several reports exist in the lit-
erature describing the influence of NNS on HR and pe-
ripheral oxygenation [17, 51–53] , studied mainly during
painful procedures  or in gavage-fed preterm infants
[17, 51, 52] , these data are not comparable to ours since
these studies were not carried out during sleep and did
not differentiate between suckling bursts and rest peri-
Franco et al.  studied the influence of NNS on HR
variability in sleeping infants. She found significant dif-
ferences not only between pacifier users and nonusers,
but also during periods with and without suckling in pac-
ifier users, suggesting that pacifier use modifies the in-
fant’s cardiac autonomic control during sleep. According
to our results, this apparently does not, however, result in
a change of mean HR.
Similarly, RF did not change significantly during
bursts in our study, while Daniels et al.  reported that
RF increased during suckling bursts when compared to
rest periods in preterm infants.
In conclusion, our study is unable to explain the po-
tentially beneficial effect of pacifiers concerning SIDS on
a physiological background. However, the results pre-
sented in our study offer important data for suckling hab-
its in pacifier users and may thus be the basis for further
investigations on this aspect.
1 Prechtl HFR: Ultrasound studies of human
fetal behaviour. Early Hum Dev 1985;
2 Turgeon-O’Brien H, Lachapelle D, Gagnon
PF, Larocque I, Maheu-Robert LF: Nutritive
and nonnutritive sucking habits: a review.
ASDC J Dent Child 1996;
3 van Woerden EE, van Geijn HP, Caron FJ,
van der Valk AW, Swartjes JM, Arts NF: Fetal
mouth movements during behavioural states
IF and 2F. Eur J Obstet Gynecol Reprod Biol
4 Goldson E: Non-nutritive sucking in the sick
infant. J Perinatol 1987;
5 Daniels H, Devlieger H, Casaer P, Callens M,
Eggermont E: Nutritive and non-nutritive
sucking in preterm infants. J Dev Physiol
6 VandenBerg KA: Nippling management of
the sick neonate in the NICU: the disorga-
nized feeder. Neonatal Netw 1990;
7 Wolff PH: The serial organization of sucking
in the young infant. Pediatrics 1968;
8 Pansy J, Zotter H, Sauseng W, Schneuber S,
Lang U, Kerbl R: Pacifier use: what makes
mothers change their mind? Acta Paediatr
9 Larsson E. Artificial sucking habits: etiolo-
gy, prevalence and effect on occlusion. Int J
Orofacial Myology 1994;
10 Jackson JM, Mourino AP: Pacifier use and
otitis media in infants twelve months of age
or younger. Pediatr Dent 1999;
11 Victora CG, Behague DP, Barros FC, Ollinto
MTA, Weiderpass E: Pacifier use and short
breastfeeding duration: cause, consequence,
or coincidence? Pediatrics 1997;
12 Righard L: Are breastfeeding problems re-
lated to incorrect breastfeeding technique
and the use of pacifiers and bottles? Birth
13 Howard C, Howard F, Lanphear B: Random-
ized clinical trial of pacifier use and bottle-
feeding or cupfeeding and their effect on
breastfeeding. Pediatrics 2003;
14 Kramer M, Barr R, Dagenais S: Pacifier use,
early weaning, and cry/fuss behavior: a ran-
domized controlled trial. JAMA 2001;
15 Collins CT, Ryan P, Crowther CA, McPhee
AJ, Paterson S, Hiller JE: Effect of bottles,
cups, and dummies on breast feeding in pre-
term infants: a randomised controlled trial.
16 Mathai S, Natrajan N, Rajalakshmi NR: A
comparative study of nonpharmacological
methods to reduce pain in neonates. Indian
17 Pinelli J, Symington A: Non-nutritive suck-
ing for promoting physiologic stability and
nutrition in preterm infants. Cochrane Da-
tabase Syst Rev 2005:CD001071.
18 Measel CP, Anderson GC: Nonnutritive
sucking during tube feedings: effect on clin-
ical course in premature infants. JOGN Nurs
19 Field T, Ignatoff E, Stringer S, Brennan J,
Greenberg R, Widmayer S, Anderson GC:
Nonnutritive sucking during tube feedings:
effects on preterm neonates in an intensive
care unit. Pediatrics 1982;
20 Fucile S, Gisel E, Lau C: Oral stimulation ac-
celerates the transition from tube to oral
feeding in preterm infants. J Pediatr 2002;
21 Pickler RH, Frankel HB, Walsh KM, Thomp-
son NM: Effects of nonnutritive sucking on
the behavioral organization and feeding per-
formance in preterm infants. Nurs Res 1996;