Received: 20 November 2017 Revised: 31 January 2018 Accepted: 7 February 2018

DOI: 10.1111/jspn.12214

ORIGINAL ARTICLE

Effects of two different feeding positions on physiological

characteristics and feeding performance of preterm infants:
A randomized controlled trial

Burcu Aykanat Girgin1 Duygu Gözen2 Güner Karatekin3

1 Faculty of Health Sciences Pediatric Nursing

Department, Çankırı Karatekin University, Abstract

Çankırı, Turkey Purpose: The aim of this randomized controlled study was to determine the effect of semielevated
2 Florence Nightingale Faculty of Nursing, Pedi-
side-lying (ESL) and semielevated supine (ESU) positions, which are used to bottle-feed preterm
atric Nursing Department, Istanbul University, infants, on their physiological characteristics and feeding performance.
Istanbul, Turkey
3 Division of Neonatology, Department of Design and Methods: The sample consisted of preterm infants who were born in the 31st gesta-
Pediatrics, Zeynep Kamil Maternity and tional week and below, and met the inclusion criteria. A randomization was provided in the sample
Children Diseases Training and Research
group with a total of 80 infants including 38 infants in the ESL (experimental) group and 42 infants
Hospital, Istanbul, Turkey
in the ESU (control) group. Both groups were compared in terms of their SpO2 values, heart rates,
Correspondence
Duygu Gözen, Istanbul University Florence and feeding performances before, during, and after the feeding. The data were obtained by using
Nightingale Faculty of Nursing Pediatric Nursing a form for infant descriptive characteristics, feeding follow-up form, a Masimo Radical-7 pulse
Department, Istanbul, Turkey.
oximeter device, and a video camera.
Email: [email protected]
Results: It was determined that the infants in the ESL group had statistically significantly higher
SpO2 values (ESL: 96.77 ± 2.51; ESU: 93.48 ± 5.63) and lower heart rates (ESL: 155.87 ± 11.18;
ESU: 164.35 ± 6.00) during the feeding compared to the infants in the ESU group (p < .05). Besides,
oxygen saturation levels of the infants in the ESU group decreased more (p < .01) and their heart
rates increased more (p < .05) during the feeding than those obtained before the feeding com-
pared to the infants in the ESL group.

Practice Implications: The ESL position has a more positive effect on oxygen saturation and heart
rate of infants and it is more effective in providing a physiological stabilization during the feeding,
compared to the ESU position. According to these results, the ESL position can be recommended
for preterm feeding.

KEYWORDS
feeding, position, preterm infant, semielevated side-lying, semielevated supine

1 INTRODUCTION developed in very preterm infants, they have difficulty in achieving

coordination among the sucking-swallowing-respiratory functions
Oral feeding is a physiologically difficult and complex task that (Callen & Pinelli, 2005; Gewolb & Vice, 2006) and thus display
requires a coordination among sucking, swallowing and respiratory physiological stress symptoms such as decrease of oxygen saturation,
functions of preterm infants (Amaizu, Shulman, Schanler, & Lau, 2008; bradycardia, and tachycardia during oral feeding (Gewolb & Vice, 2006;
Thoyre, Shaker, & Pridham, 2005). Oral feeding behavior arises from Mizuno et al., 2007). This negatively affects their feeding performances
dynamic interactions of multiple systems involved in oral feeding. and causes the decrease of sucking activity (Mizuno et al., 2007), insuf-
These systems include the oral-motor, neurological, cardio-pulmonary, ficient food intake (Lau, Smith, & Schanler, 2003; Mizuno et al., 2007),
and gastrointestinal systems. Self-organization of these systems leads and prolonged feeding duration (Pridham et al., 1998). In Als's (1986)
to a functional feeding coordination among the suck-swallow-breathe Synactive theory, it is reported that preterm infants primarily need to
pattern (Goldfield, 2007). As pulmonary functions are not sufficiently succeed in controlling their autonomic systems for oral feeding. Thus,

J Spec Pediatr Nurs. 2018;e12214. wileyonlinelibrary.com/journal/jspn 

c 2018 Wiley Periodicals, Inc. 1 of 9
https://doi.org/10.1111/jspn.12214
2 of 9 GIRGIN ET AL .

there is a need for interventions that would support physiological sta- 2 METHODS
bilization of infants during oral feeding (Clark, Kennedy, Pring, & Hird,
2007). 2.1 Design
When infants receiving breast milk are able to begin oral feed-
This study was designed as randomized controlled experimental in
ing, the preferred method is breastfeeding (Edmond & Bahl, 2006)
order to determine the effect of ESL and ESU positions, used in bottle-
and breastfed preterm infants appear more physiological stable
feeding of very preterm infants, on their physiological characteristics
than bottle-fed preterm infants (Buckley & Charles, 2006). How-
and feeding performances.
ever, preterm infants may have sucking feeds by bottle due to a
number of reasons such as infants having no mother for breast-
feeding. Family and work responsibilities make it difficult for most
2.2 Sample and setting
mothers to be present for multiple feedings every day. Bottle feed-
ing is the oral feeding method most often used when mothers Preterm infants hospitalized in the level III neonatal intensive care unit
are not present to breastfeed (McGrath, Lussier, & Briere, 2013). of a training and research hospital located in Istanbul between January
Positions to be used in very preterm infants during bottle feed- and October 2015 were included in the study. In this unit, oral feed-
ing are among important strategies for supporting physiological sta- ing initiates when infants are physiologically stable and at a mean post-
bilization (Clark et al., 2007; Park, Thoyre, Knafl, Hodges, & Nix, menstrual age of 32–33 weeks. Oral feeding is offered every 3 hours in
2014). the ESU position.
Semielevated supine position (ESU) is a position frequently used in The preterm infants meeting the following inclusion criteria were
neonatal intensive care units. In this position, infant is laid and fed on included in the study: (a) being born between 26 and 31+6 gestational
his/her back and the head is elevated to an angle of 45–60◦ in order weeks according to mother's last menstrual date, (b) being at postmen-
to reduce the respiratory load of lungs (Park et al., 2014). However, strual age of 32 to 39+6 weeks during the study and having a body
this position causes the tongue and soft palate of the infant to fall weight above 1,500 g, (c) transiting from orogastric catheter feeding
back because of gravity, thus blocks the upper airway (Litman et al., to oral feeding, (d) tolerating at least 80% of their prescribed milk dur-
2005). Also, gravity causes an abundant flow of milk from the bottle ing oral feeding for two consecutive days, (e) voluntarily signing of the
due to the effect of hydrostatic pressure occurring when the bottle written informed consent form by parents, and (f) being ready for oral
is turned down, and the infant may have difficulty in controlling the feeding according to the cue-based feeding approach. According to the
abundant flow of milk and suffer from aspiration and apnea (Park et al., cue-based feeding approach, infant should tolerate full enteral feeding,
2014). Semielevated side-lying (ESL) position is a position recently have a stable oxygen saturation and respiratory system during feed-
recommended instead of the supine position (Clark et al., 2007). In ing, be able to transit to the alert state, have the ability to lick, nuz-
the ESL position, head of the infant is elevated to an angle of 45–60◦ zle or suck nonnutritively, and root in response to touch around the
and he/she is fed in the side-lying position. In this position, the bottle mouth and lips (Thoyre, Park, Pados, & Hubbard, 2013; White & Par-
is held at a lowered angle and the flow of milk in the bottle slows with nell, 2013). Infants suffering from craniofacial abnormalities like cleft
decreased hydrostatic pressure and the decreased flow of milk allows palate, cleft lip, and facial muscle paralysis; gastrointestinal, neurolog-
the infant to control the milk better (Jackman, 2013; Park et al., 2014). ical, and genetic disorders (necrotizing enterocolitis, intracranial hem-
The ESL position may also prevent anatomically the tongue and soft orrhage, periventricular leukomalacia, hydrocephalia, down syndrome,
palate from falling back and open the airway (Litman et al., 2005). Since omphalocele, short bowel syndrome etc.); moderate and severe bron-
this position mimics the breastfeeding position, very preterm infants chopulmonary dysplasia (BPD) (Jobe & Bancalari, 2001), and patent
are supported to benefit from the advantages of breastfeeding (Park ductus arteriosus requiring surgical therapy were excluded from the
et al., 2014). study.
Recent studies have investigated the effects of positions on
physiological characteristics and feeding performances of preterm
2.3 Sample size and randomization
infants (Clark et al., 2007; Dawson et al., 2013; Park et al., 2014).
In these studies, it is observed that preterm infants have higher Clark and colleagues (2007) reported a mean (SD) difference of 3
oxygen saturation values (Clark et al., 2007), less variation in heart (4.4)% in the variable of oxygen saturation between ESL and ESU
rates (Park et al., 2014), and shorter feeding durations when they positions for averagely 3 min of feeding. In order to find a similar
are fed in the side-lying position (Dawson et al., 2013). Expressing difference with theoretical power of 80% and reliability of 95%, the
the advantages of feeding in the side-lying position, the studies have minimum sample size was determined as totally 72 infants (36 in each
been emphasized to have a smaller sample size (n = 6 infants) and to group). The infants who met the sample criteria were included in the
be conducted on the same infants via repeated measurement order ESL (experimental) group and the ESU (control) group by using the urn
(Clark et al., 2007; Dawson et al., 2013; Park et al., 2014). Thus, it is randomization method (Kundt, 2007). Two balls (white for the control
reported that there is a need for randomized controlled study results group and red for the experimental group) were put into a black bag
in larger sample groups to show the effects of ESL and ESU positions by the researcher. Any nurse who was working in the unit and had no
on physiological characteristics and feeding performances of preterm information about the groups was asked to pick a ball from the black
infants. bag. The infants were assigned to either experimental group or control
GIRGIN ET AL . 3 of 9

group according to the color of the selected ball. As we had only one 2.4.4 Video camera
researcher (first author) collecting data, we could not assign her as
A video camera (Canon Legria-HFR106) was used to calculate the feed-
blind assessor during the study. But, video recordings obtained to
ing durations of the preterm infants and arrange accurately their phys-
assess the feeding duration were evaluated by a neonatal nurse spe-
iological follow-up and feeding duration records.
cialist and the first author. The neonatal nurse specialist was blinded
to the groups to eliminate possible mistakes in observations and the
2.4.5 Bottle
possibility of taking sides to avoid bias. In the power analysis repeated
A sterile, narrow-mouth, latex-free bottle and a Dr. Brown's slow-flow
at the end of the study, the power of the study was determined as 87%
pacifier were used to feed the infants in both groups and prevent the
at a reliability of 95%.
possible changes between different types of bottles in terms of milk
flow rate during the feeding.
2.4 Outcome measures and instruments
The primary outcomes were the physiological characteristics (oxy- 2.4.6 Pacifier for preterm infants
gen saturation, heart rate) and feeding performance (feeding dura- A Jollypop preterm infant pacifier was used in infants in both groups 10
tion, feeding efficiency, and percentage of food intake). Oxygen satu- min before the oral feeding for 3 min in order to apply the nonnutritive
ration and heart rate were measured by using a pulse oximeter 2 min sucking intervention.
before the feeding, during the feeding and for 30 min after the feed-
ing. Feeding duration, feeding efficiency, and percentage of food intake 2.4.7 Chronometer and sterilizer
were evaluated via a video during the feeding and recorded after the
The chronometer of Iphone 5 (“Apple”) was used to determine accu-
feeding.
rately the period from 2 min before feeding to 30 min after feeding in
both groups. A “Weewell” steam sterilizer was used to sterilize the bot-
2.4.1 Form for infant descriptive characteristics tles and the bottle pacifiers.
The form included questions about infant's gestational week at
birth, birth weight, gender, medical diagnosis/diagnoses, postmen-
2.5 Ethical considerations
strual week during the study, body weight and state, and amount of
oxygen support of the infant during the study. An ethics committee approval and an institutional approval were
received from the Ethics Committee of Zeynep Kamil Maternity

2.4.2 Feeding follow-up form and Children's Diseases Training and Research Hospital (IRB number
15.08.2014/140). Written informed consents were also received from
The form included information about percentage of the food intake in
the families. Also, the study was registered under the number: NCT
the previous meal, amount of food ordered by the physician, feeding
02962609.
duration, amount and percentage of food intake during the study, feed-
ing efficiency and physiological changes (oxygen saturation and heart
rate) observed in the infant before, during, and after the feeding. 2.6 Data collection
Feeding performance of the preterm infants was evaluated by mea-
The first author assessed all preterm infants to check the eligibility cri-
suring the percentage of food intake, feeding efficiency, and feeding
teria and the collected data. The form for infant descriptive charac-
duration (Lau & Smith, 2011; Lau, Sheena, Shulman, & Schanler, 1997;
teristics was filled for each infant before the feeding. The infants in
Pickler, Best, Reyna, Wetzel, & Gutcher, 2005). Feeding duration was
both groups were followed up 30 min before the feeding in terms of
accepted as the time elapsed from giving the bottle to the infant until
the cues indicating their readiness for feeding according to the cue-
taking it out of his/her mouth and completing the process of feeding.
based feeding approach (Kirk, Alder, & King, 2007; Thoyre et al., 2013;
The feeding duration included short-term intervals during the feed-
White & Parnell, 2013) and it was decided to feed them when they dis-
ing to rest the infant (Thoyre & Carlson, 2003). Percentage of food
played those cues. No invasive intervention was applied to infants 30
intake signified the percentage taken by the infant from amount of food
min before the feeding in order for infants in both groups not to expe-
ordered by the physician at the end of feeding duration and the feed-
rience any fatigue. The video camera was placed near the incubator
ing was accepted as successful if the percentage of the feeding intake
before the feeding in such a way to view the infants’ head and upper
was 80% and above (Lau & Smith, 2011). Feeding efficiency was evalu-
body during the feeding. The data collection was carried out at same
ated by dividing infant's total feeding duration into the amount of food
hours of the day in order to ensure that the medium light and noise lev-
intake and calculating how much milliliter of food was taken per minute
els are similar for each infant.
(Pickler et al., 2005).
Bottles and pacifiers were sterilized and nonnutritive sucking inter-
vention with pacifier was applied to the infants in both groups 10 min
2.4.3 Pulse oximeter device before the feeding for 3 min in order to increase their wakefulness
A Masimo Radical 7 (Masimo Corparation. Irvine, CA, USA) pulse levels (Fucile, Gisel, McFarland, & Lau, 2011). The amount of breast
oximeter device was used to measure physiological characteristics of milk suggested by the physician was measured by using an injector
the infants in both groups before, during, and after the feeding. and put into bottle. The pulse oximeter probe was attached to the
4 of 9 GIRGIN ET AL .

infants’ foot and the video camera was started. The infants in both between two observers at the level of intraclass correlation coeffi-
groups were loosely swaddled in a cotton cloth and cuddled in the flex- cient = 1.000 was significant (p < .01; Şencan, 2005).
ion position by the researcher. The pulse oximeter device was started
and after it showed the first measurement values, the chronometer 2.7 Data analysis
was started. The measurements were taken 2 min before the feed-
ing when the infants were calm and physiologically stable in the arms The data were analyzed by using IBM SPSS Statistics 20 (IBM Corpo-

of the researcher. After 2 min, the chronometer was stopped and the ration, Armonk, NY, USA). Descriptive analysis was performed for all

stage of feeding was started. the variables. In order to compare the two groups in terms of demo-

The infants in both groups were fed only with breast milk and only graphic characteristics and physiological characteristics, student's t-

by the researcher in order for them not to be affected by nurses’ differ- test for normally distributed variables and Mann Whitney U test for

ent feeding techniques. The infants in the experimental group (n = 38) nonnormally distributed variables were used. While Friedman test was

were held in the ESL position. In this position, their head and trunk used in within-group comparisons of three and more follow-ups of the

were elevated to an angle of 45–60◦ with the help of a pillow prepared variables of oxygen saturation and heart rate showing no normal dis-

by the researcher from the beds that were previously used in the unit tribution, Wilcoxon Signed-Rank test was used for paired comparisons.

and infants were held in the side-lying position as in the breast-feeding In order to compare the two groups in terms of feeding performance,

position where their right ear faced the ceiling and the other ear faced student's t test for normally distributed variables and Mann Whitney U

the arms of the researcher. Their knees and hip were leaned against test for nonnormally distributed variables were used. Qualitative data

the researcher's arms and their head and neck were held at the same such as gender and medical diagnosis were compared using the Fisher's

level by the researcher; whereas, their chin was held in the flexion pos- exact test and Yates’ continuity correction test. The value of p < .05 was

ture mildly facing the floor. The researcher supported their head, neck, accepted as statistically significant (Erol, 2010).

and shoulder with her left hand and controlled the bottle with her right
hand.
The ESU position was applied to the infants in the control group 3 RESULTS
(n = 42). In this position, their head and trunk were elevated to an angle
of 45–60◦ with the help of the same pillow that was prepared by the 3.1 Sample characteristics
researcher from the beds previously used in the unit and was used in Figure 1 shows the flow chart of the study. A total of 80 preterm infants
the experimental group and they was laid in the supine position in the (38 infants in the experimental-ESL group and 42 infants in the control-
arms of the researcher. Their head and neck were held at the same ESU group) were included in the study. The baseline demographic char-
level by the researcher, whereas their chin was held in the flexion pos- acteristics of the preterm infants were not statistically significantly dif-
ture mildly facing the floor. The researcher supported their head, neck, ferent between the study groups (Table 1).
and shoulder with her left hand and controlled the bottle with her right Among totally 80 infants in the ESL and the ESU groups included in
hand. the study, 78.8% (n = 63) were diagnosed with respiratory distress syn-
During the feeding, the infants in both groups were followed up in drome, 33.8% (n = 27) sepsis, 16.3% (n = 13) mild BPD, 17.5% (n = 14)
terms of their physiological–behavioral stress and fatigue symptoms patent ductus arteriosus requiring no surgical therapy, 21.3% (n = 17)
in accordance with the cue-based feeding approach (Kirk et al., 2007; respiratory distress, and 6.3% (n = 5) second-stage intracranial hem-
Thoyre et al., 2013; White & Parnell, 2013) and the feeding was inter- orrhage. There was no statistically significant difference between the
rupted for a short time in case of these symptoms. The feeding was con- two groups in terms of the prevalence of the aforementioned diag-
tinued in case of the cues showing that they were ready to be fed again noses (p > .05).
as well as physiological stability (Ludwig & Waitzman, 2007; Thoyre It was determined that the average percentage of food intake in
et al., 2013). the previous meal was 94.38 ± 5.30% (min–max: 85–100); the average
It was paid attention for feeding duration not to exceed 30 min amount of food ordered by the physician was 31.31 ± 5.09 ml during
including short-term intervals. After the feeding, the infants in both the study (min–max: 18–45 ml) and both groups were similar in terms
groups were laid on their right side in the incubators in order to allow of the aforementioned characteristics (p > .05).
the gastric discharge (Sangers et al., 2013). Any intervention possi-
bly affecting their physiological values was prevented throughout this
3.2 Physiological characteristics
period and their physiological values were measured for 30 min. At the
thirtieth minute, physiological values recorded on the pulse oximeter There was no statistically significant difference between mean oxygen
device were computerized by using the Masimo data transfer software. saturation of the infants in the ESL group and the ESU group before the
Both the researcher and an expert neonatal nurse, who knew nothing feeding (Table 2, p = .740). On the other hand, the infants in the ESL
about the sample group to eliminate the possibility of the researcher to group had a statistically significantly higher mean oxygen saturation
have any bias, watched video camera records for the purpose of deter- compared to the infants in the ESU group during (Table 2, p = .003) and
mining the feeding duration of infants and filled the feeding follow- after the feeding (Table 2, p = .020). When comparing the difference
up form. The feeding duration was determined by averaging the dura- of change in oxygen saturation levels during and before the feeding, it
tion specified by the researcher and the neonatal nurse. Agreement was determined that mean oxygen saturation decreased as 1.31 ± 2.82
GIRGIN ET AL . 5 of 9

FIGURE 1 CONSORT flow diagram of randomized controlled trial

units in the ESL group and 4.61 ± 4.88 units in the ESU group and the the ESU group during the study (Table 4, p = .049). On the other hand,
decrease in the ESU group was significantly higher than the ESL group there was no statistically significant difference between the groups in
(Table 2, p = .001). terms of feeding efficiency rate (Table 4, p = .179) and percentage of
There was no statistically significant difference between pre- food intake during the study (Table 4, p = .501).
feeding mean heart rates of the ESL group and the ESU group (Table 3,
p = .859). On the other hand, the infants in the ESL group had statis-
tically significantly lower mean heart rate compared to infants in the 3.4 Discussion
ESU group during (Table 3, p = .001) and after the feeding (Table 3,
In the present study, physiological characteristics and feeding per-
p = .037). When comparing the difference of variation in heart rates
formance were compared between ESL and ESU groups. The infants
before and after the feeding, it was determined that mean heart rates
in both groups were similar in terms of demographic characteristics
increased as 4.71 ± 8.86 units in the ESL group and 14.02 ± 20.34 units
(Howe, Sheu, Hinojosa, Lin, & Holzman, 2007; White-Traut et al., 2013)
in the ESU group and the increase in the ESU group was significantly
and medical diagnoses (Hwang, Ma, Tseng, & Tsai, 2013; White-Traut
higher than the ESL group (Table 3, p = .045).
et al., 2013), which prevented the differences to be observed between
the groups based on the effects of these factors.
Oxygen saturation values of infants in both groups before the feed-
3.3 Feeding performance
ing showed that they were ready for a safe oral feeding (Ross & Philbin,
It was determined that the infants in the ESL group had statistically sig- 2011; Thoyre & Carlson, 2003). It was determined that there was
nificantly shorter average feeding duration compared to the infants in no statistically significant difference between mean oxygen saturation
6 of 9 GIRGIN ET AL .

TA B L E 1 Comparison of descriptive characteristics of the preterm infants (N = 80)

Characteristics ESL group (n = 38) ESU group (n = 42) Test p

Gestational age (week) Mean ± SD 30.40 ± 1.07 30.23 ± 1.09 Z: –0.923 .356a
Min–Max (Median) 27.71–31.71 (30.57) 28–31.71 (30.29)
Birth weight (g) Mean ± SD 1368.55 ± 196.42 1351.19 ± 224.52 t: 0.366 .715b
Min–Max (Median) 945–1690 (1352.5) 915–1780 (1345)
Postmenstrual age (week) Mean ± SD 34.07 ± 1.32 34.23 ± 1.39 Z: –0.893 .372a
Min–Max (Median) 32.57–37.71 (33.57) 32.43–37.86 (33.79)
Body weight (g) Mean ± SD 1767.37 ± 153.94 1783.33 ± 153.62 t: –0.464 .644b
Min–Max (Median) 1520–2140 (1750) 1505–2160 (1775)
n (%) n (%)
Gender Girl 19 (50.0) 22 (52.4) 𝜒 2 : 0.001 1.000c
Boy 19 (50.0) 20 (47.6)
a Mann Whitney U test; b Student's t-test; c Yates continuity correction test.

TA B L E 2 Comparison of oxygen saturation levels of the preterm infants (N = 80)

Oxygen saturation level (%)

Time of measurement ESL group (n = 38) ESU group (n = 42) Test p
Before feeding Mean ± SD 98.08 ± 1.61 98.09 ± 1.78 Z: –0.332 .740a
Min–Max (Median) 92.68–100 (98.48) 93.28–100 (98.37)
During feeding Mean ± SD 96.77 ± 2.51 93.48 ± 5.63 Z: –2.996 .003a
Min–Max (Median) 88–99.84 (97.48) 73.18–100 (94.60)
After feeding Mean ± SD 97.98 ± 1.72 96.72 ± 2.83 Z: –1.946 .020b
Min–Max (Median) 92.97–99.99 (98.40) 87.58–99.82 (97.49)
Test 𝜒 2 : 8.895 𝜒 2 : 35.246
c
p .012 .001
Before–during feeding Difference –1.31 ± 2.82 –4.61 ± 4.88 Z: –3.454 .001a
Test Z: –2.792 Z: –5.073
d
p .005 .001
a Mann Whitney U test; b Student's t-test; c Friedman test, d Wilcoxon signed ranks test.

values of the infants in the ESL and ESU groups before the feeding and both studies had similar postmenstrual week averages, a difference
the mean oxygen saturation levels during and after the feeding were was observed in oxygen saturation decrease level, which is associated
higher in the ESL group than the ESU group. The difference of change with the fact that infants in the present study were evaluated before
in oxygen saturation levels measured during feeding compared to the the feeding based on the criteria of cue-based feeding approach.
prefeeding period also confirms this result (Table 2). These results In their study comparing the ESL and ESU positions in six preterm
showed that the infants in the ESU group had a greater difficulty dur- infants in the 32nd–38th postmenstrual weeks, Park and colleagues
ing feeding than infants in the ESL group. Moreover, since the ESL (2014) did not report the values of decrease in oxygen saturation val-
position mimics the breast-feeding position, it is believed to provide ues during feeding according to the positions and they determined that
a better fluid flow rate, support the coordination among the sucking- the infants had a lower decrease in oxygen saturation level in the ESL
swallowing-breathing and to be effective in providing a patency of the position. However, this result was not statistically significant. When
airway. comparing the results of Park and colleagues (2014) and the results of
In their study comparing the effect of ESL and ESU positions on the present study, it was observed that the decrease in oxygen satura-
physiological findings of six preterm infants who were in the 35th– tion levels was statistically significantly higher in the ESU group in the
38th postmenstrual weeks, Clark et al. (2007) reported that oxygen present study (Table 2). The difference between the results of these
saturation levels of infants decreased from 98.28 ± 1.83 (before feed- two studies was associated with the fact that the sample size in the
ing) to 91.14 ± 4.99 (during feeding) in 3 min in the ESU position and present study was larger than the sample size in the study by Park and
from 96.44 ± 3.84 to 94.01 ± 3.37 in the ESL position. The decrease in colleagues (2014).
oxygen saturation levels of both ESL and ESU groups during the feeding When comparing mean heart rate of preterm infants before the
was observed to be lower in the present study, compared to the results feeding, it was observed that there was no statistically significant dif-
of the study by Clark and colleagues (2007). Even though infants in ference between ESL and ESU groups and mean heart rates were
GIRGIN ET AL . 7 of 9

TA B L E 3 Comparison of heart rates of the preterm infants (N = 80)

Heart rate (min)

Time of measurement ESL group (n = 38) ESU group (n = 42) Test p
Before feeding Mean ± SD 151.16 ± 14.72 150.33 ± 18.60 Z: –0.178 .859a
Min–Max (Median) 103.45–171.45 (155.68) 101.70–176.30 (150.68)
During feeding Mean ± SD 155.87 ± 11.18 164.35 ± 6.00 Z: –3.444 .001a
Min–Max (Median) 128.69–173.92 (158.87) 156.23–182.35 (162.97)
After feeding Mean ± SD 150.56 ± 10.77 155.18 ± 13.68 Z: –2.091 .037a
Min–Max (Median) 129.43–174.96 (150.31) 110.92–187.26 (156.09)
Test 𝜒 2 : 12.474 𝜒 2 : 16.714
b
p .002 .001
Before–during feeding Difference 4.71 ± 8.86 14.02 ± 20.34 Z: –2.004 .045a
Test Z: –3.154 Z: –3.920
cp .002 .001
a b c
Mann Whitney U test; Friedman test; Wilcoxon Signed Ranks test.

TA B L E 4 Comparison of feeding performances of the preterm infants (N = 80)

Feature of feeding performance ESL group (n = 38) ESU group (n = 42) Test p
Feeding duration/min Mean ± SD 13.11 ± 5.72 15.83 ± 6.83 Z: –2.092 .049a
Min–Max (Median) 7–30 (11) 5–30 (14)
Feeding efficiency (ml/min) Mean ± SD 2.68 ± 0.95 2.38 ± 1.07 t: 1.355 .179b
Min–Max (Median) 0.73–5.28 (2.72) 0.72–5.6 (2.3)
Percentage of food intake (%) Mean ± SD 99.63 ± 1.93 99.59 ± 2.12 Z: 0.673 .501a
Min–Max (Median) 88.8–100 (100) 88–100 (100)
a
Mann Whitney U test; b Student's t-test.

similar and in normal range in both groups in terms of readiness better adaptation to physiological changes, and had more regular heart
of infants for feeding (Thoyre & Carlson, 2003). In the literature, it rates during the feeding. In addition, it was observed that the preterm
is suggested that since lung functions do not sufficiently develop in infants in the ESL position had higher oxygen saturation values and
very preterm infants, they do not have a coordination among the lower heart rates during the feeding, with a statistically significant dif-
sucking-swallowing-breathing functions and thus, respiratory func- ference compared to the preterm infants in the ESU position and it was
tions are suppressed and oxygen saturation levels decrease (Mizuno thought that those results supported each other.
et al., 2007). It is known that very preterm infants increase their heart In their study, Park and colleagues (2014) reported that the heart
rates in order to organize this change experienced in respiratory func- rates of preterm infants showed less variation during the feeding in
tions and oxygen saturation levels during oral feeding, which results in the ESL position in a statistically significant level compared to infants
sending sufficient oxygen to tissues as a response to the increased res- in the ESU position (ESL: 0.04 ± 0.02; ESU: 0.07 ± 0.01). In the present
piratory effort. The increase in heart rate is accepted as an effort of study, the heart rates of infants were compared in terms of difference
infants to obtain homeostasis again against the physiological compul- of change before-during the feeding. Similar to the findings of Park and
sion that occurs as a result of the transition to oral feeding. Significant colleagues (2014), the results of the present study also showed that the
increases in heart rates signify that infants are considerably affected increase in the heart rates of the infants in the ESU group during feed-
by the physiological strain that occurs during oral feeding (Blackburn, ing was statistically significantly higher compared to the infants in the
2007). ESL group (Table 3).
It was determined that the ESL group had a statistically significantly It was determined that average feeding duration were statistically
lower mean heart rate during and after the feeding compared to the significantly shorter in the infants in the ESL group compared to the
ESU group. When comparing the difference of variation in the heart infants in the ESU group. On the other hand, no statistically significant
rates of infants measured before and during the feeding, it was deter- difference was determined between the groups in terms of feeding
mined that the increase in the ESU group (14.02 ± 20.34) was signif- efficiency and percentage of food intake (Table 4). In the study by Park
icantly higher than the ESL group (4.71 ± 8.86) (Table 3). According and colleagues (2014), they reported the feeding duration of infants as
to the literature information (Blackburn, 2007; Gewolb & Vice, 2006), 13.0 min in the ESL position and 11.6 min in the ESU position. However,
these results showed that infants in the ESL group were affected less the difference was not statistically significant. In the same study, it was
by the physiological strains that were caused by feeding, displayed a reported that feeding efficiency was 2.5 ml/min in the ESL position and
8 of 9 GIRGIN ET AL .

2.4 ml/min in the ESU position and while the infants in the ESL posi- ACKNOWLEDGMENTS
tion completed 95.2% of their foods, those in the ESU position com- We would like to thank neonatal nurses and neonatologists of
pleted 92.7%. There was no statistically significant difference between the Zeynep Kamil Maternity and Children's Diseases Training and
the positions in terms of both feeding efficiency and percentage of Research Hospital for their assistance. This study was presented as
food intake (Park et al., 2014). These results were found to be similar oral in Nutrition Session of the 6th Congress of the European Academy
with those in the present study (Table 4). However, differently from of Paediatric Societies EAPS on October 21–25, 2016, in Geneva,
the study by Park and colleagues (2014), feeding duration of infants Switzerland.
was found to be statistically significantly shorter in the ESL group com-
pared to the ESU group in the present study. This difference between
FUNDING INFORMATION
the two studies was associated with the fact that sample size in the
present study was larger and randomized controlled studies were con- This study was funded by the İstanbul University Scientific Research

ducted in two different groups. Projects as Doctorate Thesis Project with the number: 48252.

In Als's (1986) Synactive theory, it is reported that preterm infants

primarily need to succeed in controlling their autonomic systems and CONFLICT OF INTEREST
thus, their motor and behavioral status is positively affected. In the lit- The authors declare that there is no conflict of interest.
erature, it is suggested that high oxygen saturation values in preterm
infants during feeding will enhance their behaviors of integrating into
ORCID
feeding (Hill, 2002; Thoyre & Brown, 2004). On the other hand, the
Burcu Aykanat Girgin http://orcid.org/0000-0002-2601-8781
decrease of oxygen saturation values will cause preterm infants to
Duygu Gözen http://orcid.org/0000-0001-9272-3561
develop fatigue symptoms in the early period, have a weaker sucking
activity (Mizuno et al., 2007), take less amount of food, and conse-
quently have a prolonged feeding duration (Lau et al., 2003). In accor- REFERENCES
dance with literature, the present study revealed that oxygen satu- Als, H. (1986). A synactive model of neonatal behavioral organiza-
ration was higher in the ESL position and thus, feeding duration was tion: Framework for the assessment of neurobehavioral develop-
ment in the premature infant and for support of infants and par-
shorter in this position.
ents in the neonatal intensive care environment: Part I, theoretical
framework. Physical & Occupational Theraphy in Pediatrics, 6(3), 3–53.
https://doi.org/10.1080/J006v06n03_02
Amaizu, N., Shulman, R. J., Schanler, R. J., & Lau, C. (2008). Maturation of
4 CONCLUSION oral feeding skills in preterm infants. Acta Paediatrica, 97(1), 61–67.
https://doi.org/10.1111/j.1651-2227.2007.00548.x
In the present study, it was found that the ESL position was more effi- Blackburn, S. T. (2007). Maternal, fetal, and neonatal physiology: A clinical per-
cient in maintaining the normal values of both oxygen saturation and spective. St. Louis: Saunders Elseiver.
heart rate and shortening the feeding duration compared to the ESU Buckley, K. M., & Charles, E. M. (2006). Benefits and challenges of transi-
position. Using ESL position that is optimal for the physiological stabil- tioning preterm infants to at-breast feedings. International Breastfeeding
ity of preterm infants would be a relatively simple practice change to Journal, 1(13), 1–7. https://doi.org/10.1186/1746-4358-1-13

implement and could be most beneficial to infant's feeding progression. Callen, J., & Pinelli, J. (2005). A review of the literature examining the ben-
efits and challenges, incidence and duration, and barriers to breast-
feeding in preterm infants. Advances in Neonatal Care, 5(2), 72–88.
https://doi.org/10.1016/j.adnc.2004.12.003
4.1 Strengths and limitations
Clark, L., Kennedy, G., Pring, T., & Hird, M. (2007). Improving bot-
The strengths of the study were as follows: the data were obtained tle feeding in preterm infants: Investigating the elevated side-
by using objective measurement instruments, all infants were fed only lying position. Infant, 3(4), 154–158. Retrieved from https://
www.infantjournal.co.uk/journal_article.html?RecordNumber=5659
with breast milk, and it is the first study comparing the ESL and the ESU
Dawson, J. A., Myers, L. R., Moorhead, A., Jacobs, S. E., Ong, K., Salo, F., …
positions in two different groups in a randomized controlled trial and
Davis, P. G. (2013). A randomised trial of two techniques for bottle feed-
having a large sample group. The results of the study were limited with ing preterm infants. Journal of Pediatrics and Child Health, 49(6), 462–
the sample group of the present study. 466. https://doi.org/10.1111/jpc.12208
Edmond, K., & Bahl, R. (2006). Optimal feeding of low- birth-
weight ınfants. Geneva: WHO. Retrieved from https://apps.who.
4.2 How might this information affect nursing int/iris/bitstream/10665/43602/1/9789241595094 _eng.pdf
practice? Erol, H. (2010). Analyzing data with SPSS program. Adana: Nobel Publishing.
Fucile, S., Gisel, E. G., McFarland, D. H., & Lau, C. (2011). Oral and non-
The ESL position supports developmental care during feeding process
oral sensorimotor interventions enhance oral feeding performance in
of preterm infants. The ESL position positively affects the physiological preterm infants. Developmental Medicine & Child Neurology, 53(9), 829–
symptoms of preterm infants. It is recommended for neonatal nurses to 835. https://doi.org/10.1111/j.1469-8749.2011.04023.x
establish evidence-based protocols in order to apply the ESL position Gewolb, I. H., & Vice, F. L. (2006). Abnormalities in the coordination of
and cue-based feeding principles in NICU. respiration and swallow in preterm infants with bronchopulmonary
GIRGIN ET AL . 9 of 9

dysplasia. Developmental Medicine & Child Neurology, 48(7), 595–599. with weak pressures to maintain breathing during feeding. Pediatrics,
https://doi.org/10.1017/S0012162206001241 120(4), e1035–1042. https://doi.org/10.1542/peds.2006-3567
Goldfield, E. C. (2007). A dynamical systems approach to infant oral Park, J., Thoyre, S., Knafl, G. J., Hodges, E. A., & Nix, W. B. (2014). Efficacy
feeding and dysphagia: From model system to therapeutic med- of semielevated side-lying positioning during bottle-feeding of very
ical device. Ecological Psychology, 19(1), 21–48. https://doi.org/ preterm infants. The Journal of Perinatal & Neonatal Nursing, 28(1), 69–
10.1080/10407410709336949 79. https://doi.org/10.1097/JPN.0000000000000004
Hill, A. S. (2002). Toward a theory of feeding efficiency for Pickler, R. H., Best, A. M., Reyna, B. A., Wetzel, P. A., & Gutcher, G.
bottle-fed preterm infants. Journal of Theory Construction & R. (2005). Prediction of feeding performance in preterm infants.
Testing, 6(1), 75–81. Retrieved from https://search.proquest. Newborn & Infant Nursing Reviews, 5(3), 116–123. https://doi.org/
com/docview/219180159?accountid=11637 10.1053/j.nainr.2005.04.001
Howe, T. H., Sheu, C. F., Hinojosa, J., Lin, J., & Holzman, I. R. Pridham, K., Brown, R., Sondel, S., Green, C., Wedel, N. Y., & Lai, H. C. (1998).
(2007). Multiple factors related to bottle-feeding perfor- Transition time to full nipple feeding for premature infants with a his-
mance in preterm infants. Nursing Research, 56(5), 307–311. tory of lung disease. Journal of Obstetric, Gynecologic, & Neonatal Nursing,
https://doi.org/10.1097/01.NNR.0000289498.99542.dd 27(5), 533–545. https://doi.org/10.1111/j.1552-6909.1998.tb02620.x
Hwang, Y. S., Ma, M. C., Tseng, Y. M., & Tsai, W. H. (2013). Associations Ross, E. S., & Philbin, M. K. (2011). Supporting oral feeding in fragile infants:
among perinatal factors and age of achievement of full oral feed- An evidence-based method for quality bottle-feedings of preterm, ill,
ing in very preterm infants. Pediatrics & Neonatology, 54(5), 309–314. and fragile infants. The Journal of Perinatal & Neonatal Nursing, 25(4),
https://doi.org/10.1016/j.pedneo.2013.03.013 349–357. https://doi.org/10.1097/JPN.0b013e318234ac7a
Jackman, K. T. (2013). Go with the flow: Choosing a feeding system Sangers, H., de Jong, P. M., Mulder, S. E., Stigter, G. D., van den Berg, C. M., te
for infants in the neonatal intensive care unit and beyond based on Pas, A. B., & Walther, F. J. (2013). Outcomes of gastric residuals whilst
flow performance. Newborn & Infant Nursing Reviews, 13(1), 31–34. feeding preterm infants in various body positions. Journal of Neonatal
https://doi.org/10.1053/j.nainr.2012.12.003 Nursing, 19(1), 337–341. https://doi.org/10.1016/j.jnn.2012.12.003
Jobe, A. H., & Bancalari, E. (2001). Bronchopulmonary dysplasia. Ameri- Şencan, H. (2005). Reliability and validity in social and behavioural measure-
can Journal of Respiratory and Critical Care Medicine, 163(7), 1723–1729. ments. Ankara: Seçkin Publishing.
https://doi.org/10.1164/ajrccm.163.7.2011060 Thoyre, S. M., & Brown, R. L. (2004). Factors contributing to preterm infant
Kirk, A. T., Alder, S. C., & King, J. D. (2007). Cue-based oral feed- engagement during bottle feeding. Nursing Research, 53(5), 304–313.
ing clinical pathway results in earlier attainment of full oral feed- https://doi.org/10.1097/00006199-200409000-00005
ing in premature infants. Journal of Perinatology, 27(9), 572–578. Thoyre, S. M., & Carlson, J. (2003). Occurrence of oxygen desat-
https://doi.org/10.1038/sj.jp.7211791 uration events during preterm infant bottle feeding near dis-
Kundt, G. (2007). A new proposal for setting parameter values in restricted charge. Early Human Development, 72(1), 25–36. https://doi.org/
randomization methods. Methods of Information in Medicine, 46(4), 440– 10.1016/S0378-3782(03)00008-2
449. https://doi.org/10.1160/ME0398 Thoyre, S., Park, J., Pados, B., & Hubbard, C. (2013). Developing a co-
Lau, C., & Smith, E. O. (2011). A novel approach to assess oral regulated, cue-based feeding practice: The critical role of assess-
feeding skills of preterm infants. Neonatology, 100(1), 64–70. ment and reflection. Journal of Neonatal Nursing, 19(4), 139–148.
https://doi.org/10.1159/000321987 https://doi.org/10.1016/j.jnn.2013.01.002
Lau, C., Sheena, H. R., Shulman, R. J., & Schanler, R. J. (1997). Oral feed- Thoyre, S. M., Shaker, C. S., & Pridham, K. F. (2005). The early feeding
ing in low birth weight infants. Journal of Pediatrics, 130(4), 561–569. skills assessment for preterm infants. Neonatal Network, 24(3), 7–16.
https://doi.org/10.1016/S0022-3476(97)70240-3 https://doi.org/10.1891/0730-0832.24.3.7
Lau, C., Smith, E. O., & Schanler, R. J. (2003). Coordination of suck-swallow White, A., & Parnell, K. (2013). The transition from tube to full oral
and swallow respiration in preterm infants. Acta Paediatrica, 92(6), 721– feeding (breast or bottle): A cue-based developmental approach.
727. https://doi.org/10.1111/j.1651-2227.2003.tb00607.x Journal of Neonatal Nursing, 19(4), 189–197. https://doi.org/
Litman, R. S., Wake, N., Chan, L. M., McDonough, J. M., Sin, S., Mahboubi, S., 10.1016/j.jnn.2013.03.006
& Arens, R. (2005). Effects of lateral positioning on upper airway size White-Traut, R., Pham, T., Ranhin, K., Norr, K., Shapiro, N., & Yoder,
and morphology in sedated children. Anesthesiology, 103(3), 484–488. J. (2013). Exploring factors related to oral feeding progression
https://doi.org/10.1097/00000542-200509000-00009. Retrieved in premature infants. Advances in Neonatal Care, 13(4), 288–294.
from https://anesthesiology.pubs.asahq.org/article.aspx?articleid= https://doi.org/10.1016/j.jnn.2013.03.006
2026337
Ludwig, S. M., & Waitzman, K. A. (2007). Changing feeding documentation to
reflect infant-driven feeding practice. Newborn & Infant Nursing Reviews,
7(3), 155–160. https://doi.org/10.1053/j.nainr.2007.06.007 How to cite this article: Girgin BA, Gözen D, Karatekin G.
McGrath, J. M., Lussier, M. M., & Briere, C. R. (2013). Bottlefeed- Effects of two different feeding positions on physiological char-
ing the breastfed infant in the neonatal intensive care unit.
acteristics and feeding performance of preterm infants: A ran-
Newborn & Infant Nursing Reviews, 13(1), 5–6. https://doi.org/
10.1053/j.nainr.2012.12.008 domized controlled trial. J Spec Pediatr Nurs. 2018;e12214.

Mizuno, K., Nishida, Y., Taki, M., Hibino, S., Murase, M., Sakurai, M., &
https://doi.org/10.1111/jspn.12214
Itabashi, K. (2007). Infants with bronchopulmonary dysplasia suckle